# Agent Skills

Agent Skills is an open format for extending AI agent capabilities with specialized knowledge and workflows, using folders centered on a SKILL.md file.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## This Month in AWS DevOps Agent: November 2026

DevFeed: [This Month in AWS DevOps Agent: November 2026](<https://devfeed.tech/articles/this-month-in-aws-devops-agent-november-2026-146990.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/mt/this-month-in-aws-devops-agent-november-2026/>)

Author: Nereida Woo

Published: 2026-11-21T01:47:57Z

Content type: release

Language: en

Sources: [AWS Cloud Operations Blog](<https://devfeed.tech/sources/aws-cloud-operations-blog.md>)

Topics: [DevOps](<https://devfeed.tech/topics/devops.md>), [DevAIOps](<https://devfeed.tech/topics/devaiops.md>), [AWS Organizations](<https://devfeed.tech/topics/aws-organizations.md>), [Monitoring](<https://devfeed.tech/topics/monitoring.md>), [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>)

Tags: [agent-sandbox](<https://devfeed.tech/tags/agent-sandbox.md>), [amazon-bedrock-guardrails](<https://devfeed.tech/tags/amazon-bedrock-guardrails.md>), [amazon-cloudwatch](<https://devfeed.tech/tags/amazon-cloudwatch.md>), [announcements](<https://devfeed.tech/tags/announcements.md>), [aws](<https://devfeed.tech/tags/aws.md>), [aws-govcloud-us](<https://devfeed.tech/tags/aws-govcloud-us.md>), [aws-organizations](<https://devfeed.tech/tags/aws-organizations.md>), [devops](<https://devfeed.tech/tags/devops.md>), [devops-agent](<https://devfeed.tech/tags/devops-agent.md>), [faster](<https://devfeed.tech/tags/faster.md>), [november-2026](<https://devfeed.tech/tags/november-2026.md>), [post-types](<https://devfeed.tech/tags/post-types.md>), [real-time](<https://devfeed.tech/tags/real-time.md>), [reviews](<https://devfeed.tech/tags/reviews.md>)

### AI overview

November updates extend AWS DevOps Agent to AWS GovCloud (US-West) and let teams associate an Agent Space with an organizational unit or organization root; organization mode is unavailable in GovCloud, and GovCloud omits some features, including Sandbox and private connections. Investigations now finish faster and stream summaries, while Sandbox (preview) runs code in an isolated environment scoped to one investigation or conversation, with read-only AWS calls and restricted network access. Teams can replay investigations to compare configuration changes, receive sampled automatic quality Reviews, and track nine CloudWatch metrics. Memory additions include importing custom knowledge, semantic search, linked memories, and weekly reviews that can remove stale claims, merge duplicates, or resolve contradictions with recorded evidence and version history. The agent also adds partner-hosted MCP connectors, a skills catalog, chat in Slack, Teams, and MCP clients, and code change risk analysis in preview. Availability and capabilities vary by Region; partner connector requests may leave the Agent Space's AWS Region.

### Source excerpt

Introduction November brought AWS DevOps Agent to AWS GovCloud (US-West) and added AWS Organizations support, so teams can connect an entire organizational unit to an Agent Space instead of one account at a time. Investigations now complete faster and show a real-time summary as they run. Replay and automatic Reviews give teams a way to [...]

## How Feedback-Free Self-Evolution Reshapes Agent Skills: A Systematic Study

DevFeed: [How Feedback-Free Self-Evolution Reshapes Agent Skills: A Systematic Study](<https://devfeed.tech/articles/how-feedback-free-self-evolution-reshapes-agent-skills-a-systematic-study-145532.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25534>)

Author: Litao Guo, Man Chen, Xinli Xu, Jinsong Zhou, Zixin Zhang, Ying-Cong Chen

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-skills](<https://devfeed.tech/tags/agent-skills.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [evolution](<https://devfeed.tech/tags/evolution.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [experts](<https://devfeed.tech/tags/experts.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>), [model](<https://devfeed.tech/tags/model.md>), [pass](<https://devfeed.tech/tags/pass.md>), [skills](<https://devfeed.tech/tags/skills.md>), [study](<https://devfeed.tech/tags/study.md>)

### AI overview

In experiments where models repeatedly rewrote expert skills without tasks, runs, or judgments, the skills became less useful: every tested model performed worse on every benchmark, with an average 15-point drop on SkillsBench after 30 rounds. The authors find that rewriting pulls skills toward what each model already tends to produce, discarding knowledge outside that prior first. Restoring this knowledge recovers much of the lost pass rate. Their SkillDiff method marks that knowledge and improves SkillsBench pass rate by an average 4.8 points, without feedback or changing the skill text. These findings are reported in the supplied abstract.

### Source excerpt

arXiv:2611.25534v1 Announce Type: new Abstract: Agent skills are documents in which experts write down procedural knowledge, and models increasingly maintain them by rewriting them without feedback. Such rewriting could adapt a skill to the model that uses it or erode the expert knowledge that makes it useful, yet how skills evolve under this feedback-free self-evolution has not been studied. We let models rewrite expert skills round after round without any task, run or judgment and run agents built on the same models with the rewritten skills. Our experiments show that the skills lose value rather than improve, as every model we test performs worse with them on every benchmark, by 15 points on average on SkillsBench after thirty rounds. Sampling each model's prior with skills it writes from the topic alone, we find that rewriting pulls skills toward this prior and gives up first the knowledge outside it. Putting this knowledge back recovers much of the lost pass rate, so an expert's skill is valuable largely for what the model does not write on its own. Building on this insight, SkillDiff marks this knowledge in place and raises the pass rate on SkillsBench by 4.8 points on average without feedback or any change to the skill's text. Project repository: https://github.com/LitaoGuo/How-Self-Evolution-Reshapes-Skills.

## Evolve Where It Matters: Coverage-Guided Self-Evolution of Agent Skills

DevFeed: [Evolve Where It Matters: Coverage-Guided Self-Evolution of Agent Skills](<https://devfeed.tech/articles/evolve-where-it-matters-coverage-guided-self-evolution-of-agent-skills-145513.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25429>)

Author: Boyin Tan, Jiadong Zhang, Xiaosong Ma, Chun Jason Xue, Xiaowei Huang, Youcheng Sun

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>)

Tags: [agent-skills](<https://devfeed.tech/tags/agent-skills.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [coverage](<https://devfeed.tech/tags/coverage.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [evolution](<https://devfeed.tech/tags/evolution.md>), [follow](<https://devfeed.tech/tags/follow.md>), [gpt-5-5](<https://devfeed.tech/tags/gpt-5-5.md>), [root-cause-analysis](<https://devfeed.tech/tags/root-cause-analysis.md>), [skills](<https://devfeed.tech/tags/skills.md>)

### AI overview

SkillPlumber uses instruction-level coverage to distinguish incorrect skill instructions from instructions an agent failed to follow: each instruction is labeled as followed, not followed, or not covered. A persistent root-cause agent turns those labels into revision questions and checks changes against the next run; a maintainer distills findings across tasks into a shared strategy library. Across three benchmarks and three executors, it solves 64.7%-70.5% of tasks, 25.6-30.4 percentage points above the initial skill and 8.7-14.0 points above the strongest baseline. On SkillsBench with GPT-5.5, removing coverage reduces recovered failures by 20.8%.

### Source excerpt

arXiv:2611.25429v1 Announce Type: cross Abstract: Agent skills package procedural knowledge as natural-language instructions, and a skill helps only when its instructions are correct and properly followed by the agent. Self-evolving agents revise their own skills from the runs they make, but existing methods judge each run as a whole, by its outcome or by where it went wrong, and cannot tell whether a failure comes from a wrong instruction or from one the agent did not follow. Instruction-level coverage was introduced to test skills and labels each instruction of a run as follow, not-follow, or not covered, which separates these causes. We leverage it for skill self-evolution with SkillPlumber, a framework in which an agent evolves its own skills where coverage shows they fall short. A persistent root cause analysis agent turns each label into a revision question and checks each revision against the next run, while a maintainer distills investigations across tasks into a shared strategy library. Across three benchmarks and three executors, SkillPlumber solves 64.7% to 70.5% of the tasks, 25.6 to 30.4 percentage points higher than the initial skill and 8.7 to 14.0 percentage points higher than the strongest baseline, and removing coverage reduces recovered failures by 20.8% on SkillsBench with GPT-5.5.

## Designer-RSI: Evolving Procedural Memory for Graphic Design Agent

DevFeed: [Designer-RSI: Evolving Procedural Memory for Graphic Design Agent](<https://devfeed.tech/articles/designer-rsi-evolving-procedural-memory-for-graphic-design-agent-145375.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2609.22086>)

Author: Hongyang Du, Lan Yan, Christian Flores, Asim Kadav

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [Tool](<https://devfeed.tech/topics/tool.md>), [Ground truth / benchmark quality](<https://devfeed.tech/topics/ground-truth-benchmark-quality.md>), [Retrieval Augmented Generation (RAG)](<https://devfeed.tech/topics/retrieval-augmented-generation-rag.md>)

Tags: [acquisition](<https://devfeed.tech/tags/acquisition.md>), [agent](<https://devfeed.tech/tags/agent.md>), [creative-software](<https://devfeed.tech/tags/creative-software.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [design](<https://devfeed.tech/tags/design.md>), [evolution](<https://devfeed.tech/tags/evolution.md>), [execution](<https://devfeed.tech/tags/execution.md>), [external](<https://devfeed.tech/tags/external.md>), [graphic-design](<https://devfeed.tech/tags/graphic-design.md>), [memory](<https://devfeed.tech/tags/memory.md>), [modular](<https://devfeed.tech/tags/modular.md>), [procedural](<https://devfeed.tech/tags/procedural.md>), [rsi](<https://devfeed.tech/tags/rsi.md>), [skills](<https://devfeed.tech/tags/skills.md>), [vlm](<https://devfeed.tech/tags/vlm.md>)

### AI overview

Designer-RSI adapts a graphic design agent with frozen foundation-model weights by evolving a modular skill bank from unlabeled design trajectories. It widens coverage by adding skills for recurring gaps and deepens execution quality by revising failing skills against successful runs. A Pareto-style matched replay gate accepts updates only when they improve sampled briefs without local regressions, addressing noisy visual judges and costly, confounded tool runs. Across held-out product briefs, eight external benchmarks, and two backbones, the approach outperformed static documentation skills, with execution success reported up to 99.8% and blind win rate up to 67.6% over the base agent. Ablations indicate acquisition and revision work synergistically; neither alone provided reliable gains.

### Source excerpt

arXiv:2609.22086v2 Announce Type: replace Abstract: Deployed software agents face a continuous stream of user requests while their foundation-model weights remain frozen between releases. Recursive self-improvement (RSI) via external memory offers a practical path to continual adaptation, yet prior methods largely assume programmatic verifiers, cheap retries, or a single global prompt. Creative software breaks these assumptions, as open-ended visual briefs lack ground truth, VLM judges are noisy, and multi-step tool executions are costly and confounded by external assets. We present Designer-RSI, which evolves a modular SKILL.md skill bank from unlabeled design trajectories for an agent spanning over 230 raster, vector, and layout tools. Designer-RSI couples widening (acquiring new skills for recurring capability gaps) with deepening (revising failing skills by contrasting successful and failed runs), and screens updates through a Pareto-style matched replay gate that requires strict gains on sampled briefs without local regressions. Across held-out product briefs, eight external benchmarks, and two backbones, traffic-driven evolution outperforms static documentation skills, lifts execution success up to 99.8%, improves requirement fulfillment, and reaches up to a 67.6% blind win rate over the base agent. Ablations further show that skill acquisition and revision are strongly synergistic: neither suffices alone, and reliable gains require evolving coverage and execution quality together.

## SKILL.PNG: Learning Agent Skills in Pixel Space

DevFeed: [SKILL.PNG: Learning Agent Skills in Pixel Space](<https://devfeed.tech/articles/skill-png-learning-agent-skills-in-pixel-space-145160.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.26296>)

Author: Maximilian Seeth, Chengzhi Martin Hu, Marie Kempkes, Paul R\\"ottger, Leo Schwinn

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [vlm](<https://devfeed.tech/topics/vlm.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Model Development](<https://devfeed.tech/topics/model-development.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [gradient-descent](<https://devfeed.tech/tags/gradient-descent.md>), [hybrid](<https://devfeed.tech/tags/hybrid.md>), [image](<https://devfeed.tech/tags/image.md>), [instructions](<https://devfeed.tech/tags/instructions.md>), [language](<https://devfeed.tech/tags/language.md>), [learning](<https://devfeed.tech/tags/learning.md>), [llms](<https://devfeed.tech/tags/llms.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [optimized](<https://devfeed.tech/tags/optimized.md>), [performance](<https://devfeed.tech/tags/performance.md>), [pixel](<https://devfeed.tech/tags/pixel.md>), [pixels](<https://devfeed.tech/tags/pixels.md>), [skill](<https://devfeed.tech/tags/skill.md>), [vision-language-models](<https://devfeed.tech/tags/vision-language-models.md>), [vision-language-models-vlms](<https://devfeed.tech/tags/vision-language-models-vlms.md>), [vlms](<https://devfeed.tech/tags/vlms.md>)

### AI overview

Skill.png represents agent skills as images whose pixels are optimized by gradient descent through the vision encoder of a frozen vision-language model (VLM). The authors report that these image skills beat frontier-LLM-optimized textual skills in 9 of 16 settings across four VLMs and four benchmarks, using 64-256 image tokens at inference versus about 900-2,000 text tokens. Jointly optimized text and image skills further improve performance. An image optimized across an ensemble can transfer among its VLMs, while the reported compatibility claim is for VLMs. The abstract describes the evaluation results but does not provide per-setting details.

### Source excerpt

arXiv:2611.26296v1 Announce Type: new Abstract: AI agents are commonly specialized for new tasks and domains using skills, i.e. natural language instructions placed in their context. However, the discrete nature of textual skills prevents continuous optimization. Gradient-based approaches such as finetuning, on the other hand, are typically model-specific, preventing skill transfer across different agents. Since many agents today are built on vision-language models (VLMs), we can address these limitations by introducing Skill.png: a method for obtaining skill-like image inputs whose pixels are optimized with gradient descent through the vision encoder of a frozen VLM. Image skills are compatible with any VLM, and a single image can be optimized jointly for an ensemble of models. We show that Skill.png significantly outperforms state-of-the-art textual skills optimized using frontier LLMs in 9 of 16 settings (across 4 VLMs and 4 benchmarks) while requiring only 64 to 256 image tokens instead of roughly 900 to 2000 text tokens at inference time. Hybrid skills, i.e. text and image skills jointly optimized, further improve agent performance. Moreover, an image optimized with an ensemble of VLMs can successfully transfer across the models in the ensemble. Skill.png thus presents the first agent skill that can be optimized with gradient descent while retaining transfer benefits of text skills. Our code is available at: https://github.com/omseeth/skill.png

## Behavioral Integrity Verification for AI Agent Skills

DevFeed: [Behavioral Integrity Verification for AI Agent Skills](<https://devfeed.tech/articles/behavioral-integrity-verification-for-ai-agent-skills-145331.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2605.11770>)

Author: Yuhao Wu, Tung-Ling Li, Hongliang Liu

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [ai security](<https://devfeed.tech/topics/ai-security.md>), [threat detection](<https://devfeed.tech/topics/threat-detection.md>)

Tags: [adversarial](<https://devfeed.tech/tags/adversarial.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agent-skills](<https://devfeed.tech/tags/agent-skills.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [ai-agent-skills](<https://devfeed.tech/tags/ai-agent-skills.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [attack](<https://devfeed.tech/tags/attack.md>), [attack-chains](<https://devfeed.tech/tags/attack-chains.md>), [auditing](<https://devfeed.tech/tags/auditing.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [capabilities](<https://devfeed.tech/tags/capabilities.md>), [capability](<https://devfeed.tech/tags/capability.md>), [categories](<https://devfeed.tech/tags/categories.md>), [cause](<https://devfeed.tech/tags/cause.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-sy](<https://devfeed.tech/tags/cs-sy.md>), [eess-sy](<https://devfeed.tech/tags/eess-sy.md>), [integrity](<https://devfeed.tech/tags/integrity.md>), [skills](<https://devfeed.tech/tags/skills.md>), [verification](<https://devfeed.tech/tags/verification.md>)

### AI overview

Behavioral integrity verification (BIV) checks whether an AI agent skill's declared capabilities match what its code and instructions enable. It pairs deterministic code analysis with LLM-assisted capability extraction over a shared taxonomy, producing evidence for deviation analysis, root-cause classification, and malicious-skill detection. In 49,943 OpenClaw skills, 80.0% had at least one classified deviation; the authors attributed 81.1% of deviations to developer oversight and 18.9% to adversarial intent, and predicted multi-stage attack chains in 5.0% of skills. On a separate 906-skill benchmark, BIV achieved an F1 of 0.949, outperforming the reported rule-based and single-pass LLM baselines. These are results reported in the supplied abstract.

### Source excerpt

arXiv:2605.11770v2 Announce Type: replace-cross Abstract: Agent skills extend LLM agents with privileged third-party capabilities such as filesystem access, credentials, network calls, and shell execution. Existing safety work catches malicious prompts and risky runtime actions, but the skill artifact itself goes unverified. We formalize this as the behavioral integrity verification (BIV) problem: a typed set comparison between declared and actual capabilities over a shared taxonomy that bridges code, instructions, and metadata. The BIV framework instantiates this comparison by pairing deterministic code analysis with LLM-assisted capability extraction. The resulting structured evidence supports three downstream analyses: deviation taxonomy, root-cause classification, and malicious-skill detection. On 49,943 skills from the OpenClaw registry, the deviation taxonomy reveals a pervasive description-implementation gap: 80.0% of skills have at least one deviation in the taxonomy, with four compound-threat categories surfaced. Root-cause classification finds that deviations are mostly oversight, not malice: 81.1% trace to developer oversight and 18.9% to adversarial intent, with 5.0% of skills carrying predicted multi-stage attack chains. On a 906-skill malicious-skill detection benchmark, BIV reaches an F1 of 0.949, outperforming state-of-the-art rule-based and single-pass LLM baselines. These results demonstrate behavioral integrity auditing for agent skills at scale.

## EvoSkill Injection: Red-Teaming Autonomous Skill Generation and Evolution in Self-Evolving Agents

DevFeed: [EvoSkill Injection: Red-Teaming Autonomous Skill Generation and Evolution in Self-Evolving Agents](<https://devfeed.tech/articles/evoskill-injection-red-teaming-autonomous-skill-generation-and-evolution-in-self-evolving-agents-145367.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2608.30429>)

Author: Doyun Kim, Chanwoo Kim, Sugyeong Eo, Yeo-Chan Yoon, Chanjun Park

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [attack](<https://devfeed.tech/tags/attack.md>), [attack-surface](<https://devfeed.tech/tags/attack-surface.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [behaviors](<https://devfeed.tech/tags/behaviors.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [benchmark-dataset](<https://devfeed.tech/tags/benchmark-dataset.md>), [capabilities](<https://devfeed.tech/tags/capabilities.md>), [corruption](<https://devfeed.tech/tags/corruption.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [evolution](<https://devfeed.tech/tags/evolution.md>), [framework](<https://devfeed.tech/tags/framework.md>), [generation](<https://devfeed.tech/tags/generation.md>), [interaction](<https://devfeed.tech/tags/interaction.md>), [llm](<https://devfeed.tech/tags/llm.md>), [malicious](<https://devfeed.tech/tags/malicious.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [red-teaming](<https://devfeed.tech/tags/red-teaming.md>), [skill](<https://devfeed.tech/tags/skill.md>)

### AI overview

The paper defines EvoSkill Injection as a threat to self-evolving agents that generate, store, and reuse skills: malicious capabilities can be formed and treated as legitimate skills. It introduces SARGE, a red-teaming framework using iterative generation, escalation, and reinforcement, plus EvoSkillBench for malicious interaction trajectories and EvoSkillSafetyBench for assessing whether injected skills are later retrieved and activated harmfully. The reported evaluation found that SARGE induced malicious skill formation and that injected skills were persistently stored and repeatedly activated. The supplied text is an abstract, so it does not provide further evaluation details.

### Source excerpt

arXiv:2608.30429v2 Announce Type: replace Abstract: LLM-based agent systems increasingly adopt skill-based architectures to reduce repetitive reasoning costs and improve stable, efficient task execution. Recent studies propose self-evolving agents that autonomously generate, refine, and reuse skills from past experiences to enable continuous capability evolution. However, autonomous skill evolution introduces a new attack surface in which malicious capabilities are generated, stored, and reused as legitimate skills. In this paper, we define EvoSkill Injection as a threat model targeting the autonomous skill generation and evolution pipeline of self-evolving agents. We further propose SARGE (Red-teaming Autonomous Skill Generation and Evolution in self-evolving agents), a red-teaming framework for evaluating this threat model through iterative generation, escalation, and reinforcement interactions. To support our framework, we construct EvoSkillBench, a benchmark dataset of malicious interaction trajectories for inducing malicious skill formation in self-evolving agents, and introduce EvoSkillSafetyBench, a post-attack benchmark for evaluating whether injected malicious skills are subsequently retrieved and activated as harmful behaviors. Our evaluation shows that SARGE induces malicious skill formation and that injected skills are persistently stored and repeatedly activated, highlighting the risk of persistent capability corruption.

## LLM assisted reverse engineering

DevFeed: [LLM assisted reverse engineering](<https://devfeed.tech/articles/llm-assisted-reverse-engineering-144490.md>)

Original publisher: [Read original article](<https://blog.adafruit.com/2026/11/19/llm-assisted-reverse-engineering/>)

Author: Anne Barela

Published: 2026-11-19T22:17:09Z

Content type: news

Language: en

Sources: [Adafruit Industries](<https://devfeed.tech/sources/adafruit-industries-makers-hackers-artists-designers-and-engineers.md>)

Topics: [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [Ghidra](<https://devfeed.tech/topics/ghidra.md>), [code analysis](<https://devfeed.tech/topics/code-analysis.md>), [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>)

Tags: [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-coding-agent](<https://devfeed.tech/tags/ai-coding-agent.md>), [assembly](<https://devfeed.tech/tags/assembly.md>), [csv](<https://devfeed.tech/tags/csv.md>), [debugging](<https://devfeed.tech/tags/debugging.md>), [decompilation](<https://devfeed.tech/tags/decompilation.md>), [ghidra](<https://devfeed.tech/tags/ghidra.md>), [github](<https://devfeed.tech/tags/github.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [large-language-models](<https://devfeed.tech/tags/large-language-models.md>), [log](<https://devfeed.tech/tags/log.md>), [markdown-files](<https://devfeed.tech/tags/markdown-files.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [programming](<https://devfeed.tech/tags/programming.md>), [radare2](<https://devfeed.tech/tags/radare2.md>), [reverse-engineering](<https://devfeed.tech/tags/reverse-engineering.md>), [skills](<https://devfeed.tech/tags/skills.md>), [software](<https://devfeed.tech/tags/software.md>)

### AI overview

The author uses an AI coding agent to analyze binaries and produce code overviews that would take much longer to assemble by hand from assembly. One published skill documents tools such as radare2 and a Ghidra MCP bridge, with 22 scripts for targeted decompilation; it addresses agents stopping early because they do not know how to run disassembly. A second skill makes a stuck agent state one testable hypothesis at a time and log results in a CSV loop, helping prevent random guessing and circular work. Findings are recorded in Markdown as the analysis proceeds.

### Source excerpt

ioma8 has been taking apart hardware and software for two decades, and in a post on slowchronicle.com describes how that work now runs with an AI coding agent doing the reading. The agent is pointed at the binary, along with a set of purpose-written skills, and it produces an overview of the code that would [...]

## Skill Roulette: Measuring Skill Invocation Rate Across LLM Coding Agents

DevFeed: [Skill Roulette: Measuring Skill Invocation Rate Across LLM Coding Agents](<https://devfeed.tech/articles/skill-roulette-measuring-skill-invocation-rate-across-llm-coding-agents-142141.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23266>)

Author: Jin Rae Lee, Jai Lal Lulla, Sebastian Baltes, Christoph Treude

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [LLM evaluation / benchmarking](<https://devfeed.tech/topics/llm-evaluation-benchmarking.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-harnesses](<https://devfeed.tech/tags/agent-harnesses.md>), [agent-skills](<https://devfeed.tech/tags/agent-skills.md>), [agents](<https://devfeed.tech/tags/agents.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [coding](<https://devfeed.tech/tags/coding.md>), [coding-agent](<https://devfeed.tech/tags/coding-agent.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [harness-configuration](<https://devfeed.tech/tags/harness-configuration.md>), [instructions](<https://devfeed.tech/tags/instructions.md>), [language-model](<https://devfeed.tech/tags/language-model.md>), [large-language-model](<https://devfeed.tech/tags/large-language-model.md>), [llm](<https://devfeed.tech/tags/llm.md>), [measuring](<https://devfeed.tech/tags/measuring.md>), [metadata](<https://devfeed.tech/tags/metadata.md>), [models](<https://devfeed.tech/tags/models.md>), [repository](<https://devfeed.tech/tags/repository.md>), [skill](<https://devfeed.tech/tags/skill.md>), [skills](<https://devfeed.tech/tags/skills.md>), [software-engineering](<https://devfeed.tech/tags/software-engineering.md>)

### AI overview

The study measures how often LLM coding agents invoke a relevant skill when given identical tasks, and tests factors that may affect invocation. It plans 20,400 trials across four software engineering categories in a synthetic repository, using four agent harnesses and two models. Nine non-directional hypotheses cover skill metadata, competing skills, contextual instructions, task wording and composition, plus whether skill availability changes an unrequested action. Comparisons are prespecified and evaluated within each model and harness configuration using Boschloo's exact test. The supplied text describes the planned methodology, not results.

### Source excerpt

arXiv:2611.23266v1 Announce Type: new Abstract: Agent skills package procedures for large language model (LLM) coding agents. An agent selects a skill from its name and description before reading the procedure. It may complete the task without invoking a relevant skill. We investigate how reliably coding agents invoke relevant skills and which factors affect that decision. We estimate invocation rates by repeating trials with identical inputs; invocation rate is the proportion of trials in which the agent invokes the relevant skill. We test 9 non-directional hypotheses about skill metadata, competing skills, contextual instructions, and task wording and composition. Eight concern invocation; the ninth tests whether making a skill available changes the occurrence of an action that the task does not request. We use tasks from 4 software engineering categories in a synthetic repository, running each of 4 coding-agent harnesses with 2 models. The planned 20,400 trials include 60 repetitions per Baseline category and configuration, 60 more per category in Claude Code with Sonnet 5.5 that admit each category's task and skill, and 10 per experimental condition. Comparisons are specified before collection and evaluated within each model and harness configuration using Boschloo's exact test.

## SCAFFOLD: Toward Warrant-Preserving Evidence Transport for Reusable LLM-Agent Skills

DevFeed: [SCAFFOLD: Toward Warrant-Preserving Evidence Transport for Reusable LLM-Agent Skills](<https://devfeed.tech/articles/scaffold-toward-warrant-preserving-evidence-transport-for-reusable-llm-agent-skills-142204.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23990>)

Author: Nami Cho, Shailja Gupta

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [authority to operate](<https://devfeed.tech/topics/authority-to-operate.md>), [audit](<https://devfeed.tech/topics/audit.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-skill](<https://devfeed.tech/tags/agent-skill.md>), [agent-skills](<https://devfeed.tech/tags/agent-skills.md>), [authorization](<https://devfeed.tech/tags/authorization.md>), [context](<https://devfeed.tech/tags/context.md>), [contracts](<https://devfeed.tech/tags/contracts.md>), [coverage](<https://devfeed.tech/tags/coverage.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ma](<https://devfeed.tech/tags/cs-ma.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [dependencies](<https://devfeed.tech/tags/dependencies.md>), [effectiveness](<https://devfeed.tech/tags/effectiveness.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [evaluations](<https://devfeed.tech/tags/evaluations.md>), [llm](<https://devfeed.tech/tags/llm.md>), [maintenance](<https://devfeed.tech/tags/maintenance.md>), [oracles](<https://devfeed.tech/tags/oracles.md>), [permissions](<https://devfeed.tech/tags/permissions.md>), [policies](<https://devfeed.tech/tags/policies.md>), [policy](<https://devfeed.tech/tags/policy.md>), [skills](<https://devfeed.tech/tags/skills.md>), [transport](<https://devfeed.tech/tags/transport.md>), [warrant](<https://devfeed.tech/tags/warrant.md>)

### AI overview

The paper proposes Warranted Reuse Claims (WRCs) as defeasible, claim-specific authorizations for reusing a versioned LLM-agent skill in a target context. Its Warrant-Preserving Evidence Transport (WPET) discipline asks which historical evidence remains admissible after context changes, while distinguishing admissible evidence from proof that a safety or effectiveness claim is true. The proposal includes three-valued applicability, asymmetric context-transfer judgments, and attention to provenance dependence and policy-invalidating changes. It states a conditional preservation result with limits and proposes a longitudinal comparison against version-aware baselines at matched coverage, utility, and review cost. This is a formal problem statement and set of testable obligations, not an implemented verifier, empirical superiority result, or proof of universal agent safety.

### Source excerpt

arXiv:2611.23990v1 Announce Type: cross Abstract: An agent skill that succeeded yesterday may be executable today without being justified for reuse today. Tests may cover only observable contract clauses, outcome oracles may be absent, evaluations may inherit the same erroneous specification, and permissions or dependencies may change between validation and invocation. We argue that the appropriate unit of prospective trust is a Warranted Reuse Claim (WRC): a defeasible, claim-specific authorization tied to a versioned skill, target context, evidence, assumptions, and identified defeaters. The paper sharpens this position through Warrant-Preserving Evidence Transport (WPET), a proposed discipline for deciding which historical evidence remains admissible after context change. It distinguishes evidence admissibility from truth of the supported safety or effectiveness claim; states a conditional preservation result and its limits; specifies three-valued applicability and asymmetric context-transfer judgments; and illustrates how provenance dependence and policy-invalidating changes alter authorization. We explicitly relate the proposal to assurance cases, contracts, and skill-library maintenance rather than claiming novelty for their ingredients. A falsifiable longitudinal evaluation would compare WRC-guided policies with strong version-aware baselines at matched coverage, utility, and review cost. This paper contributes a formalized problem statement, testable design obligations, and reporting protocol, not an implemented verifier, empirical superiority result, or proof of universal agent safety.

## automodel: A Skill for Agentic Model Discovery

DevFeed: [automodel: A Skill for Agentic Model Discovery](<https://devfeed.tech/articles/automodel-a-skill-for-agentic-model-discovery-142781.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24347>)

Author: Marco Virgolin, Harry Desmond, Ester Bergantin, Giorgio Morales, John W. Sheppard, Luigi Romano, Alessandro Lucantonio

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Claude](<https://devfeed.tech/topics/claude.md>), [codex](<https://devfeed.tech/topics/codex.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-skill](<https://devfeed.tech/tags/agent-skill.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agriculture](<https://devfeed.tech/tags/agriculture.md>), [algorithms](<https://devfeed.tech/tags/algorithms.md>), [astro-ph-co](<https://devfeed.tech/tags/astro-ph-co.md>), [astro-ph-ga](<https://devfeed.tech/tags/astro-ph-ga.md>), [astrophysics](<https://devfeed.tech/tags/astrophysics.md>), [automated](<https://devfeed.tech/tags/automated.md>), [automation](<https://devfeed.tech/tags/automation.md>), [case-studies](<https://devfeed.tech/tags/case-studies.md>), [claude](<https://devfeed.tech/tags/claude.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [code](<https://devfeed.tech/tags/code.md>), [codex](<https://devfeed.tech/tags/codex.md>), [coding](<https://devfeed.tech/tags/coding.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ne](<https://devfeed.tech/tags/cs-ne.md>), [data-driven](<https://devfeed.tech/tags/data-driven.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [domain](<https://devfeed.tech/tags/domain.md>), [model](<https://devfeed.tech/tags/model.md>), [skill](<https://devfeed.tech/tags/skill.md>)

### AI overview

automodel packages a model-discovery methodology as a portable skill for general-purpose coding agents. In case studies spanning astrophysics, cardiovascular dynamics, precision agriculture, traffic flow, and vehicle tire dynamics, the authors report that commercial agents using the skill identified novel, scientifically relevant models. They present this approach as a simple, interactive, broadly applicable alternative to domain-specific automated discovery pipelines; the supplied abstract does not provide details for comparing performance or limitations across the case studies.

### Source excerpt

arXiv:2611.24347v1 Announce Type: new Abstract: The search for better models, whether expressed as analytical equations or neural architectures, has increasingly shifted from manual design toward data-driven automation. While a multitude of algorithms exist, ranging from exhaustive search to AlphaEvolve, recent advances in general-purpose coding agents such as Claude Code and OpenAI Codex raise the question: could providing such systems with a "model discovery recipe" be enough to discover or improve models? We present automodel, an agent skill for general model discovery. With case studies in astrophysics, cardiovascular dynamics, precision agriculture, traffic flow, and vehicle tire dynamics, we show how commercial agents equipped with automodel can successfully identify novel and scientifically-relevant models. Our findings suggest that encoding model-discovery methodology as a portable agent skill can provide a simple, interactive, powerful and broadly applicable alternative to domain-specific automated discovery pipelines. The skill is available at: https://github.com/unlayer-ai/automodel.

## Guarding the Kingdom of Skill-Based Agents in the AI Dynasty of Computation: Control-Flow Integrity for Agent Skills

DevFeed: [Guarding the Kingdom of Skill-Based Agents in the AI Dynasty of Computation: Control-Flow Integrity for Agent Skills](<https://devfeed.tech/articles/guarding-the-kingdom-of-skill-based-agents-in-the-ai-dynasty-of-computation-control-flow-integrity-for-agent-skills-143145.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24185>)

Author: Bonan Ruan, Yuhao Sun, Jiahao Liu, Qixiao Lin, Jian Mao, Zhenkai Liang

Published: 2026-11-19T05:00:00Z

Content type: news

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [SELinux](<https://devfeed.tech/topics/selinux.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [about](<https://devfeed.tech/tags/about.md>), [abstract](<https://devfeed.tech/tags/abstract.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agent-skills](<https://devfeed.tech/tags/agent-skills.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [all](<https://devfeed.tech/tags/all.md>), [also](<https://devfeed.tech/tags/also.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [based](<https://devfeed.tech/tags/based.md>), [blocks](<https://devfeed.tech/tags/blocks.md>), [boundary](<https://devfeed.tech/tags/boundary.md>), [but](<https://devfeed.tech/tags/but.md>), [call](<https://devfeed.tech/tags/call.md>), [control](<https://devfeed.tech/tags/control.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [flow](<https://devfeed.tech/tags/flow.md>), [integrity](<https://devfeed.tech/tags/integrity.md>), [skill](<https://devfeed.tech/tags/skill.md>), [skills](<https://devfeed.tech/tags/skills.md>)

### AI overview

SkillCFI applies control-flow integrity to procedural agent skills by compiling a skill into agent instructions and an executable contract, then admitting tool requests only when trusted evidence supports the next step. It rejects skills whose mandatory obligations cannot be checked across the tool boundary and refuses unmediated channels such as inline scripts. In 770 sessions, it blocked all 240 tested violations before dispatch that a permission allowlist admitted; normal task completion was nearly unchanged. Refusals listing legal next steps increased recovery from 21% to 65%. Checking took about 2% of task time, while a compact command form reduced session slowdown from 1.32x to 1.08x. None of 53 public skills was accepted as written, though diagnostics guided seven toward enforceable contracts. These results are reported for the paper's tested skills and sessions.

### Source excerpt

arXiv:2611.24185v1 Announce Type: new Abstract: Control-flow integrity (CFI) confines a program to a control-flow graph fixed ahead of time. Procedural agent skills move this control flow into prompt text, and a large language model (LLM) agent decides at runtime whether each step runs in its reasoning, a tool call, or a script. The procedure's control skeleton survives, but its enforceability depends on where each step runs. Many violations, such as overwriting a file before validation, use only operations the skill requires, so permission allowlists admit them. A deviating or prompt-injected model can also evade a monitor by regime migration, for example by claiming that a check passed. We present SkillCFI, which carries CFI over to skill-based agents. Its compiler confines LLM interpretation to compile time and derives the agent's instructions and an executable contract from one skill. By a formal enforceability criterion, it enforces obligations whose evidence crosses the tool boundary and rejects a skill whose mandatory obligation leaves nothing to check. A deterministic runtime enforcer admits a tool request only on trusted evidence and refuses unmediated channels such as inline scripts, so admitted traces follow the contract whatever the model proposes. On eight self-written skills, no compiled contract permits more than its source under 24 source mutations. In 770 agent sessions, SkillCFI blocks before dispatch all 240 such violations, which an allowlist admits, and normal tasks complete nearly as often as without it. Listing the next legal steps in a refusal raises recovery from 21% to 65%. Checking takes about 2% of task time, and a compact command form cuts session slowdown from 1.32 times to 1.08 times. None of 53 public skills is accepted as written. The compiler's diagnostics point to interfaces rather than workflows and guided seven into enforceable contracts.

## 131 Agent Skills That Plan, Debug and Review Code Like a Staff Engineer

DevFeed: [131 Agent Skills That Plan, Debug and Review Code Like a Staff Engineer](<https://devfeed.tech/articles/131-agent-skills-that-plan-debug-and-review-code-like-a-staff-engineer-140883.md>)

Original publisher: [Read original article](<https://hackernoon.com/131-agent-skills-that-plan-debug-and-review-code-like-a-staff-engineer?source=rss>)

Author: Code Review

Published: 2026-11-18T15:30:04Z

Content type: release

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-skills](<https://devfeed.tech/tags/agent-skills.md>), [ai-coding](<https://devfeed.tech/tags/ai-coding.md>), [anthropic](<https://devfeed.tech/tags/anthropic.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [build](<https://devfeed.tech/tags/build.md>), [builds](<https://devfeed.tech/tags/builds.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [cli](<https://devfeed.tech/tags/cli.md>), [cloudflare](<https://devfeed.tech/tags/cloudflare.md>), [code](<https://devfeed.tech/tags/code.md>), [code-review](<https://devfeed.tech/tags/code-review.md>), [code-review-agent](<https://devfeed.tech/tags/code-review-agent.md>), [code-review-skills](<https://devfeed.tech/tags/code-review-skills.md>), [codex](<https://devfeed.tech/tags/codex.md>), [codex-cli](<https://devfeed.tech/tags/codex-cli.md>), [coding-agent](<https://devfeed.tech/tags/coding-agent.md>), [debug](<https://devfeed.tech/tags/debug.md>), [debugging](<https://devfeed.tech/tags/debugging.md>), [debugging-agents](<https://devfeed.tech/tags/debugging-agents.md>), [debugging-skills](<https://devfeed.tech/tags/debugging-skills.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [jesse-vincent](<https://devfeed.tech/tags/jesse-vincent.md>), [plan](<https://devfeed.tech/tags/plan.md>), [review](<https://devfeed.tech/tags/review.md>), [skills](<https://devfeed.tech/tags/skills.md>)

### AI overview

The Dev Workflow plugin packages 131 open-licensed agent skills for planning, debugging, review, architecture, refactoring, and deployment. The excerpt distinguishes the collections: Tech Leads Club focuses on architecture, deploys, Nx, performance, and plan-to-implementation; Matt Pocock's skills interview users and turn answers into specs, tickets, and code; Cloudflare's cover its platform; Thomas Ricouard's focus on Apple development plus parallel review and debugging; Superpowers emphasizes brainstorming, planning, testing, and verification; Anthropic contributes four skills. Skills load when relevant and use the open agentskills.io format, supported by 32+ tools, including Claude Code and Codex CLI. The excerpt says all listed skills are MIT- or Apache-licensed and retain attribution. It gives a Cuttlefish marketplace install command, or a manual repo-copy option for compatible agents; Codex CLI reads project `.agents/skills/` or global `~/.agents/skills/`.

### Source excerpt

One command now gives a coding agent 131 open-licensed engineering skills from Tech Leads Club, Matt Pocock, Cloudflare, Thomas Ricouard, Jesse Vincent and Anth

## SPFx Dev Skills for SPFx 1.24.1: tested across agents and models

DevFeed: [SPFx Dev Skills for SPFx 1.24.1: tested across agents and models](<https://devfeed.tech/articles/spfx-dev-skills-for-spfx-1-24-1-tested-across-agents-and-models-140557.md>)

Original publisher: [Read original article](<https://devblogs.microsoft.com/microsoft365dev/spfx-dev-skills-for-spfx-1-24-1-tested-across-agents-and-models/>)

Author: Vesa Juvonen, Waldek Mastykarz

Published: 2026-11-18T11:56:00Z

Content type: release

Language: en

Sources: [Microsoft 365 Developer](<https://devfeed.tech/sources/microsoft-365-developer-blog.md>)

Topics: [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [development-process](<https://devfeed.tech/topics/development-process.md>), [Visual Studio Code](<https://devfeed.tech/topics/visual-studio-code.md>), [Claude](<https://devfeed.tech/topics/claude.md>)

Tags: [ai-coding-agents](<https://devfeed.tech/tags/ai-coding-agents.md>), [api](<https://devfeed.tech/tags/api.md>), [apis](<https://devfeed.tech/tags/apis.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [claude-opus-5-5](<https://devfeed.tech/tags/claude-opus-5-5.md>), [claude-sonnet-5-5](<https://devfeed.tech/tags/claude-sonnet-5-5.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [copilot](<https://devfeed.tech/tags/copilot.md>), [copilot-chat](<https://devfeed.tech/tags/copilot-chat.md>), [copilot-fusion](<https://devfeed.tech/tags/copilot-fusion.md>), [dependencies](<https://devfeed.tech/tags/dependencies.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [microsoft-365](<https://devfeed.tech/tags/microsoft-365.md>), [microsoft-365-copilot](<https://devfeed.tech/tags/microsoft-365-copilot.md>), [microsoft-365-developer](<https://devfeed.tech/tags/microsoft-365-developer.md>), [microsoft-copilot](<https://devfeed.tech/tags/microsoft-copilot.md>), [microsoft-teams](<https://devfeed.tech/tags/microsoft-teams.md>), [migration](<https://devfeed.tech/tags/migration.md>), [models](<https://devfeed.tech/tags/models.md>), [react](<https://devfeed.tech/tags/react.md>), [sharepoint](<https://devfeed.tech/tags/sharepoint.md>), [sharepoint-framework](<https://devfeed.tech/tags/sharepoint-framework.md>), [skills](<https://devfeed.tech/tags/skills.md>), [spfx](<https://devfeed.tech/tags/spfx.md>), [upgrade](<https://devfeed.tech/tags/upgrade.md>)

### AI overview

SPFx Dev Skills v1.1.2 adds SPFx 1.24.1 project creation and upgrades, React 18 migration guidance, current tooling requirements, complete CLI for Microsoft 365 upgrade report handling, and updated Fluent UI v9 and PnPjs recommendations. In 105 Windows runs across seven agent-model combinations, its rubric scores improved for every combination: six averaged 100%, while GPT-6.1 Terra in Copilot Chat averaged 98%. Results varied in cost: GPT-6 Luna and Copilot Fusion each averaged 100% at $0.03 and $0.02 per run, compared with $0.82 for Copilot Chat using Opus 5.5. Fusion's result cannot be attributed to one model. The evaluation covered only one scenario--upgrading an existing project--and does not establish that smaller models or routing will suffice for other SPFx tasks.

### Source excerpt

SPFx Dev Skills v1.1.2 adds support for SPFx v1.24.1 and React 18 migration. Across 105 evaluation runs, the updated skill improved upgrade quality for every agent and model combination we tested, with 6 of 7 reaching a 100% mean score. The post SPFx Dev Skills for SPFx 1.24.1: tested across agents and models appeared first on Microsoft 365 Developer Blog.

## Identifying and Repairing Energy Smells with Agent Skills

DevFeed: [Identifying and Repairing Energy Smells with Agent Skills](<https://devfeed.tech/articles/identifying-and-repairing-energy-smells-with-agent-skills-139619.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23203>)

Author: Mohammadjavad Mehditabar, Saurabhsingh Rajput, Tushar Sharma

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [Green Software](<https://devfeed.tech/topics/green-software.md>), [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [Static code analysis](<https://devfeed.tech/topics/static-code-analysis.md>)

Tags: [agent-skills](<https://devfeed.tech/tags/agent-skills.md>), [computer-science](<https://devfeed.tech/tags/computer-science.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [energy](<https://devfeed.tech/tags/energy.md>), [energy-efficiency](<https://devfeed.tech/tags/energy-efficiency.md>), [python](<https://devfeed.tech/tags/python.md>)

### AI overview

DR. GREENS brings energy-smell checks into Python developers' coding environment, pairing a taxonomy of 12 smells and 65 root causes with repair guidance from 230 measured cases. It combines a static analyzer with an independent verifier to check that attempted repairs preserve behavior and save energy. Across 18 sessions on three open-source repositories, using two models and three configurations, 68.6% of 258 measured repairs saved energy, with a median saving of 24.9%. The full skill found more energy smells with verified savings than a baseline without the taxonomy across all three repositories and both models. These results are limited to the reported evaluation.

### Source excerpt

arXiv:2611.23203v1 Announce Type: new Abstract: Energy smells are inefficient implementations that waste energy. Once code reaches production, these smells become costlier and riskier to fix, so energy efficiency checks should shift left, i.e., move earlier in the development lifecycle, to where code is written. Existing energy smell tools rely on fixed static rules and do not measure energy savings, whereas LLM-based optimizers mostly target execution time and rarely ground their changes in concrete smell definitions. To address both limitations, we present DR. GREENS, an agent skill that detects and repairs energy smells in Python code within the developer's coding environment. The skill grounds the coding agent in our taxonomy of 12 energy smells and 65 root causes, augmented with repair guidance drawn from 230 measured cases. The skill also encapsulates a static analyzer and an independent verifier to confirm behavior preservation and energy saving for each attempted repair. We evaluated DR. GREENS in 18 sessions on three open-source repositories, with two models and three skill configurations. Of the 258 measured repairs, 68.6% saved energy, with a median saving of 24.9%. Compared with a baseline skill that lacks the taxonomy, the full skill found more energy smells with verified savings in all three repositories with both models. DR. GREENS helps developers detect energy hotspots and fix them reliably without altering program behavior, which makes energy-efficient development more feasible.

## When Agent Skills Go Stale: Runtime Behavior and Selective Maintenance in Evolving Environments

DevFeed: [When Agent Skills Go Stale: Runtime Behavior and Selective Maintenance in Evolving Environments](<https://devfeed.tech/articles/when-agent-skills-go-stale-runtime-behavior-and-selective-maintenance-in-evolving-environments-140043.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22525>)

Author: \\"Omer Faruk Akg\\"ul, Shuli Jiang, Yuting Zhang, Yi Zhang, Xiaohan Fei, Wei Xia, Stefano Soatto

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-skills](<https://devfeed.tech/tags/agent-skills.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [environments](<https://devfeed.tech/tags/environments.md>), [execution](<https://devfeed.tech/tags/execution.md>), [maintenance](<https://devfeed.tech/tags/maintenance.md>), [performance](<https://devfeed.tech/tags/performance.md>), [reuse](<https://devfeed.tech/tags/reuse.md>), [runtime](<https://devfeed.tech/tags/runtime.md>), [skills](<https://devfeed.tech/tags/skills.md>), [state](<https://devfeed.tech/tags/state.md>)

### AI overview

Across four task settings and two models, reusing skills after environment changes sometimes reduced performance by over 10 percentage points or raised execution cost by nearly 40% compared with using no skill; success could also stay constant while savings declined. Effects depended on which stored details became outdated and whether current observations could compensate. The proposed selective maintenance replays checks from successful executions without additional model calls, repairs skill content that conflicts with observed differences, validates updates, and rebuilds when local repair is insufficient. Compared with reusing stale state, maintenance improved accuracy by up to 67 percentage points and reduced execution steps by up to 37%. The excerpt does not specify the changes tested or further study details.

### Source excerpt

arXiv:2611.22525v1 Announce Type: new Abstract: Agent skills use past experience and current observations to guide decisions and actions on related tasks. This experience is retained in skill state, whose reuse can amortize inference-time reasoning and environment interaction. When environment changes, parts of this state can become outdated while others remain useful. Across four task settings and two models, we compare skill reuse before and after environment changes. Previously beneficial skills can reduce task performance by over 10 percentage points or increase execution cost by nearly 40% relative to no skill. Task success can also remain unchanged while execution savings decline. These effects depend on which environment-specific details become outdated and the agent's ability to compensate using current observations. We introduce selective maintenance, which replays checks from successful executions without additional model calls. It verifies whether observed differences conflict with skill state, repairs affected content, and validates updates, rebuilding when local repair is insufficient. No description of the change is supplied. Relative to reusing stale state, maintenance improves task accuracy by up to 67pp and reduces execution steps by up to 37%. These results suggest that preserving a skill's inference time value as environments evolve requires selectively maintaining its state.

## SkillBuffBench: Measuring What Agent Skills Contribute across Harnesses, Models, Skill Access Modes and Memory

DevFeed: [SkillBuffBench: Measuring What Agent Skills Contribute across Harnesses, Models, Skill Access Modes and Memory](<https://devfeed.tech/articles/skillbuffbench-measuring-what-agent-skills-contribute-across-harnesses-models-skill-access-modes-and-memory-139946.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22017>)

Author: Gengyue Dong, Ye Wei, Zheng Zhu, Yidong Han, Jiale Miao, Xiaojun Jia, Guanghui Li

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [Factuality scoring in AI](<https://devfeed.tech/topics/factuality-scoring-in-ai.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-harnesses](<https://devfeed.tech/tags/agent-harnesses.md>), [agent-skills](<https://devfeed.tech/tags/agent-skills.md>), [anthropic](<https://devfeed.tech/tags/anthropic.md>), [automatic](<https://devfeed.tech/tags/automatic.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [biology](<https://devfeed.tech/tags/biology.md>), [bootstrap](<https://devfeed.tech/tags/bootstrap.md>), [coding](<https://devfeed.tech/tags/coding.md>), [coding-agent](<https://devfeed.tech/tags/coding-agent.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [design](<https://devfeed.tech/tags/design.md>), [domain](<https://devfeed.tech/tags/domain.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [experts](<https://devfeed.tech/tags/experts.md>), [google](<https://devfeed.tech/tags/google.md>), [measuring](<https://devfeed.tech/tags/measuring.md>), [memory](<https://devfeed.tech/tags/memory.md>), [models](<https://devfeed.tech/tags/models.md>), [skills](<https://devfeed.tech/tags/skills.md>)

### AI overview

SkillBuffBench is designed to measure how agent Skills affect research-task performance across different harnesses, models, Skill access modes, and memory settings. It compares no Skill, a curated Skill, and automatic invocation from an installed library, with memory off or on; each model is run under all four harnesses. Current expert-contributed tasks and Skills cover materials science and biology. A deterministic verifier scores each run, paired task-level bootstrap estimates uncertainty, and cost, agent rounds, and wall-clock time are reported separately from scores. The paper describes the benchmark and protocol; evaluation is still in progress, so it reports no results yet. The task suite and analysis are intended to expand over time.

### Source excerpt

arXiv:2611.22017v1 Announce Type: new Abstract: Agent Skills, file-based packages of procedural knowledge that a coding agent can discover and load on demand, are now shipped by every major agent harness, yet the evidence for how much they help comes almost entirely from settings in which the right Skill is hand-picked for the task and the agent starts from a blank slate. Neither holds in deployment, where an agent chooses from whatever library happens to be installed and frequently carries memory from earlier work. We present SkillBuffBench, a benchmark and paired evaluation protocol that treats Skill access as a three-level factor (no Skill, a curated Skill, and automatic invocation from an installed library), crosses it with memory (off and on), with four agent harnesses and with multiple models (the state-of-the-art proprietary models from Anthropic, Google and OpenAI, and the state-of-the-art open-weight models from Chinese providers), every model being run under every harness, and evaluates the result on research tasks contributed by domain experts together with the Skills they use, currently in materials science and biology, with the suite extended on a rolling basis. Every run is marked by a deterministic verifier; uncertainty is reported with a paired, task-level bootstrap; and cost, agent rounds and wall-clock time are reported alongside the score rather than folded into it. This version of the paper describes the design of the benchmark, its evaluation protocol and its current tasks and Skills. The evaluation is in progress: results will be added once the runs are complete, and the task suite, results and analysis will continue to be updated on a rolling basis.

## WebAgents: Unbundling Agent Creation, Hosting, and Execution for an Agent-Native Web

DevFeed: [WebAgents: Unbundling Agent Creation, Hosting, and Execution for an Agent-Native Web](<https://devfeed.tech/articles/webagents-unbundling-agent-creation-hosting-and-execution-for-an-agent-native-web-139941.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.21978>)

Author: Dario Fenoglio, Martin Gjoreski, Marc Langheinrich, Pradyumna Chari, Ramesh Raskar

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [Agent Browser](<https://devfeed.tech/topics/agent-browser.md>), [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [Cryptography](<https://devfeed.tech/topics/cryptography.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-behavior](<https://devfeed.tech/tags/agent-behavior.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agents](<https://devfeed.tech/tags/agents.md>), [api](<https://devfeed.tech/tags/api.md>), [authorization](<https://devfeed.tech/tags/authorization.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [autonomous-agents](<https://devfeed.tech/tags/autonomous-agents.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [execution](<https://devfeed.tech/tags/execution.md>), [hosting](<https://devfeed.tech/tags/hosting.md>), [large-language-models](<https://devfeed.tech/tags/large-language-models.md>), [literature](<https://devfeed.tech/tags/literature.md>), [live](<https://devfeed.tech/tags/live.md>), [local](<https://devfeed.tech/tags/local.md>), [models](<https://devfeed.tech/tags/models.md>), [openclaw](<https://devfeed.tech/tags/openclaw.md>), [package](<https://devfeed.tech/tags/package.md>), [packages](<https://devfeed.tech/tags/packages.md>), [permissions](<https://devfeed.tech/tags/permissions.md>), [web](<https://devfeed.tech/tags/web.md>)

### AI overview

WebAgents separates agent creation, hosting, and execution using a signed package of portable workflows, skills, schemas, and resource requirements. An OpenClaw-based browser/runtime verifies the package, obtains user authorization, and connects it at run time to user-controlled models, tools, permissions, and budgets; hosts distribute the public program and may keep private services server-side. The authors report three workflows running unchanged with local and API-backed models. In a controlled suite, the system blocked 716 injected contract violations and accepted 295 matched benign controls, with median automated pre-execution processing below 0.5 seconds for each package. The evidence here is limited to the abstract and page excerpt.

### Source excerpt

arXiv:2611.21978v1 Announce Type: new Abstract: Large language models are moving from isolated assistants toward autonomous agents that can plan, use tools, and coordinate with others, motivating an emerging Agentic Web of independently developed agents and services. Realizing this vision requires agent behavior to move across creators, hosts, and runtimes without coupling publication to a particular execution environment or transferring authority over user resources. We introduce WebAgents, which unbundles these roles through a signed agent.pkg that carries portable workflows, skills, schemas, and resource requirements. An OpenClaw-based agent-native browser/runtime verifies the package, obtains authorization, and binds it at run time to user-controlled models, tools, permissions, and budgets, while hosts distribute the public program and may retain private services server-side. We demonstrate this contract with three unrelated workflows executed unchanged under local and API-backed models, including literature review with live services and a website-published travel agent using semantic tools and user-selected personal resources. In a controlled suite, WebAgents explicitly blocks all 716 injected contract violations while accepting all 295 matched benign controls, with median automated pre-execution processing below 0.5 s for all three packages. We conclude by outlining a path toward NANDA-based discovery and interoperable package ecosystems while retaining execution authority in the user's runtime.

## EvoSkill-S1: Evolving Reusable Skills with Genetic-Inspired Operators

DevFeed: [EvoSkill-S1: Evolving Reusable Skills with Genetic-Inspired Operators](<https://devfeed.tech/articles/evoskill-s1-evolving-reusable-skills-with-genetic-inspired-operators-140057.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22610>)

Author: Yujun Wang, Jiqiang Li, Yinuo Lu, Jinhe Bi, Aniri, Wenxuan Ye, Boliang Liu, Sikuan Yan, Shuning Wang, Xuebing Zhou, S{\\"o}ren Pirk, Hinrich Sch{\\"u}tze, Yunpu Ma

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>), [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [code](<https://devfeed.tech/tags/code.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [evolution](<https://devfeed.tech/tags/evolution.md>), [fusion](<https://devfeed.tech/tags/fusion.md>), [gpt](<https://devfeed.tech/tags/gpt.md>), [improvement](<https://devfeed.tech/tags/improvement.md>), [interface](<https://devfeed.tech/tags/interface.md>), [math](<https://devfeed.tech/tags/math.md>), [mutation](<https://devfeed.tech/tags/mutation.md>), [operators](<https://devfeed.tech/tags/operators.md>), [procedural](<https://devfeed.tech/tags/procedural.md>), [qwen](<https://devfeed.tech/tags/qwen.md>), [recovery](<https://devfeed.tech/tags/recovery.md>), [reflection](<https://devfeed.tech/tags/reflection.md>), [revision](<https://devfeed.tech/tags/revision.md>), [rl](<https://devfeed.tech/tags/rl.md>), [skills](<https://devfeed.tech/tags/skills.md>), [train](<https://devfeed.tech/tags/train.md>)

### AI overview

EvoSkill-S1 evolves reusable skills directly, using reinforcement learning to train an evolver to turn skills linked to hard-negative failures into candidate updates through five operators: revision, cross-revision, mutation, dual-reflection, and crossover. Downstream task rewards guide generation, and evaluator-guided write-back selectively retains beneficial updates. Skill Bank Fusion combines operator-specific banks, while a Procedural Skill Interface is intended to support transfer to unseen problems. The abstract reports improvements over initialized skill banks across all reported math and code benchmarks with Qwen and GPT-family solvers, reaching up to 12.39% relative gain; against the strongest external baseline without direct skill evolution, the largest reported relative improvement is 20.39%.

### Source excerpt

arXiv:2611.22610v1 Announce Type: new Abstract: Reusable skills allow language-model agents to accumulate and reuse problem-solving experience. However, most existing methods improve a skill bank only indirectly through rollout-driven accumulation and heuristic reflection, often resulting in shallow and weakly controlled updates. We present EvoSkill-S1, an RL-driven framework that instead treats reusable skills themselves as the direct objects of evolution. Given hard-negative failures, EvoSkill-S1 selects parent skills and transforms them into candidate child skills through five genetic-inspired operators: revision, cross-revision, mutation, dual-reflection, and crossover. Downstream task rewards train the skill evolver to generate more effective child skills, while evaluator-guided write-back selectively retains beneficial updates. Complementary Skill Bank Fusion consolidates operator-specific banks to exploit their distinct recovery patterns, while a Procedural Skill Interface supports transfer to unseen problems. Across all reported math and code benchmarks, EvoSkill-S1 improves the corresponding initialized skill banks under both Qwen and GPT-family solvers, with relative gains of up to 12.39%. Compared with the strongest external baseline without direct skill evolution, the largest relative improvement reaches 20.39%.

## Agentic engineering with GitLab and your local agent: A walkthrough

DevFeed: [Agentic engineering with GitLab and your local agent: A walkthrough](<https://devfeed.tech/articles/agentic-engineering-with-gitlab-and-your-local-agent-a-walkthrough-144773.md>)

Original publisher: [Read original article](<https://about.gitlab.com/blog/agentic-engineering-walkthrough/>)

Author: Fatima Sarah Khalid

Published: 2026-11-18T00:00:00Z

Content type: tutorial

Language: en

Sources: [GitLab](<https://devfeed.tech/sources/gitlab.md>)

Topics: [GitLab](<https://devfeed.tech/topics/gitlab.md>), [agentic-coding](<https://devfeed.tech/topics/agentic-coding.md>), [Claude Code](<https://devfeed.tech/topics/claude-code.md>), [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [gitlab-ci](<https://devfeed.tech/topics/gitlab-ci.md>), [code security](<https://devfeed.tech/topics/code-security.md>), [Loop Engineering](<https://devfeed.tech/topics/loop-engineering.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>), [codex](<https://devfeed.tech/topics/codex.md>), [cursor](<https://devfeed.tech/topics/cursor.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-engineering](<https://devfeed.tech/tags/agentic-engineering.md>), [agentic-workflows](<https://devfeed.tech/tags/agentic-workflows.md>), [ai](<https://devfeed.tech/tags/ai.md>), [code](<https://devfeed.tech/tags/code.md>), [code-reviews](<https://devfeed.tech/tags/code-reviews.md>), [codex](<https://devfeed.tech/tags/codex.md>), [coding](<https://devfeed.tech/tags/coding.md>), [connect](<https://devfeed.tech/tags/connect.md>), [cursor](<https://devfeed.tech/tags/cursor.md>), [developer](<https://devfeed.tech/tags/developer.md>), [devsecops](<https://devfeed.tech/tags/devsecops.md>), [duo](<https://devfeed.tech/tags/duo.md>), [environment](<https://devfeed.tech/tags/environment.md>), [failed](<https://devfeed.tech/tags/failed.md>), [flow](<https://devfeed.tech/tags/flow.md>), [gitlab](<https://devfeed.tech/tags/gitlab.md>), [harness](<https://devfeed.tech/tags/harness.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [install](<https://devfeed.tech/tags/install.md>), [issue](<https://devfeed.tech/tags/issue.md>), [lifecycle](<https://devfeed.tech/tags/lifecycle.md>)

### AI overview

The GitLab Duo setup skill connects a local coding harness to GitLab Duo Agent Platform, where repository-aware work runs as GitLab flows and merge requests remain subject to existing review, pipeline, security, and approval policies. After reviewing one setup plan, an Owner can enable group workflows, automatic code review, failed-pipeline automation, and the GitLab MCP connection; a project Maintainer can configure only that project. The walkthrough covers handing assigned issues to Developer, reviewing MRs and resolving feedback or conflicts, fixing failed pipelines, and--on Ultimate--triaging security findings. Pipeline fixes appear as inline suggestions when they fit the MR diff, as a new MR when they require other files or no MR is open, or as guidance when a reliable fix cannot be determined or the failure is security-sensitive. Prerequisites include a GitLab.com group with GitLab Credits, a project with pipelines and work items, and Claude Code plus git and glab; group-wide setup requires the Owner role. The pipeline flow reads only the last 150 KiB of a job log. Security automation offers standard and conservative presets with different thresholds and open-MR limits.

### Source excerpt

You can build an agentic engineering environment to drive greater automation throughout your workflow. How? Start in the harness you already use and connect to GitLab Duo Agent Platform once with a skill. Longer, repository-aware workflows run on GitLab, and every merge request still hits the same reviews, pipelines, scans, and approval policies. In this article, we will show you how to quickly get started, from setup to tackling four key use cases. We focus on Claude Code as the local harness because that's what we see most of our customers using but this same flow can apply to the local harness of your choosing, including Duo CLI. How to configure your agentic engineering environment If you already do most of your coding in a local harness, keep working there and wire it to GitLab Duo Agent Platform. From the terminal, you can direct lifecycle work to agentic workflows that run inside GitLab: Picking up an issue and opening a merge request Reviewing every MR as it lands Resolving merge conflicts Repairing a pipeline that just failed With GitLab Ultimate, triaging and fixing security findings once code lands Claude Code, Codex, Cursor, OpenCode, and others become the place from where you direct that work. Part 1 is the setup Install the GitLab Duo setup skill, ask Claude Code to run it, and it sets up your whole top-level group with the necessary settings, the flows, an automation in every project, and the MCP connection. Part 2 walks through four workflows You can try them as soon as the setup is done, each driven from Claude Code: Find an issue assigned to you and hand it to the Developer flow. Get automated code reviews on your merge requests, act on the findings without leaving your harness, and resolve any conflicts that pop up. Use an automation to fix a failed pipeline. On Ultimate, triage and fix security findings with the security flows. When you want an even tighter loop The MR followup skill guides your local harness through the steps above even faster.

## Reducing database incident MTTR by 73% with a custom AWS DevOps Agent Skill

DevFeed: [Reducing database incident MTTR by 73% with a custom AWS DevOps Agent Skill](<https://devfeed.tech/articles/reducing-database-incident-mttr-by-73-with-a-custom-aws-devops-agent-skill-139360.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/devops/reducing-database-incident-mttr-by-73-with-a-custom-aws-devops-agent-skill/>)

Author: Chetan Suri

Published: 2026-11-17T22:27:48Z

Content type: tutorial

Language: en

Sources: [AWS DevOps & Developer Productivity Blog](<https://devfeed.tech/sources/aws-devops-developer-productivity-blog.md>)

Topics: [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [Amazon Aurora](<https://devfeed.tech/topics/amazon-aurora.md>), [DevOps](<https://devfeed.tech/topics/devops.md>), [Monitoring](<https://devfeed.tech/topics/monitoring.md>), [Post Mortem](<https://devfeed.tech/topics/post-mortem.md>)

Tags: [agent-skill](<https://devfeed.tech/tags/agent-skill.md>), [aws](<https://devfeed.tech/tags/aws.md>), [best-practices](<https://devfeed.tech/tags/best-practices.md>), [database](<https://devfeed.tech/tags/database.md>), [decision](<https://devfeed.tech/tags/decision.md>), [devops](<https://devfeed.tech/tags/devops.md>), [devops-agent](<https://devfeed.tech/tags/devops-agent.md>), [incidents](<https://devfeed.tech/tags/incidents.md>), [intermediate-200](<https://devfeed.tech/tags/intermediate-200.md>), [mttr](<https://devfeed.tech/tags/mttr.md>), [postgresql](<https://devfeed.tech/tags/postgresql.md>), [reduction](<https://devfeed.tech/tags/reduction.md>)

### AI overview

A custom AWS DevOps Agent Skill uses signal classification and a prioritized decision tree to guide Aurora PostgreSQL incident investigations across five failure modes: connection exhaustion, replication lag, slow query regression, storage pressure, and parameter drift. In a controlled internal test of three simulated incidents, average time to identify root cause fell from 52 to 14 minutes, and wrong hypotheses per incident fell from 3.0 to 0.3. The comparison covered connection exhaustion, slow query regression, and parameter drift, so the reported results do not establish performance across every branch or production conditions. The guide requires an AWS account and Agent Space access to the cluster, CloudWatch metrics and Performance Insights, and the Operator Web App; it recommends deploying for Incident RCA and On-demand agents, excluding Incident Triage because the procedure is too thorough for fast triage.

### Source excerpt

Database incidents routinely take 45 to 60 minutes to triage. This post shows how to build a custom AWS DevOps Agent Skill that encodes expert Aurora PostgreSQL triage into decision-tree logic across five failure modes, cutting mean time to resolution from 52 minutes to 14 minutes (a 73% reduction) in a controlled test.

## Announcing AI data preparation capabilities in Amazon SageMaker: Build high-quality datasets for model development workflows

DevFeed: [Announcing AI data preparation capabilities in Amazon SageMaker: Build high-quality datasets for model development workflows](<https://devfeed.tech/articles/announcing-ai-data-preparation-capabilities-in-amazon-sagemaker-build-high-quality-datasets-for-model-development-workflows-138418.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/machine-learning/announcing-ai-data-preparation-capabilities-in-amazon-sagemaker-build-high-quality-datasets-for-model-development-workflows/>)

Author: Varun Rajan

Published: 2026-11-17T19:40:01Z

Content type: release

Language: en

Sources: [Artificial Intelligence](<https://devfeed.tech/sources/artificial-intelligence.md>)

Topics: [Amazon SageMaker](<https://devfeed.tech/topics/amazon-sagemaker.md>), [Dataset development](<https://devfeed.tech/topics/dataset-development.md>), [Data Quality](<https://devfeed.tech/topics/data-quality.md>), [AI-Ready Data](<https://devfeed.tech/topics/ai-ready-data.md>), [human feedback](<https://devfeed.tech/topics/human-feedback.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Amazon S3](<https://devfeed.tech/topics/amazon-s3.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-skill](<https://devfeed.tech/tags/agent-skill.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-data](<https://devfeed.tech/tags/ai-data.md>), [amazon](<https://devfeed.tech/tags/amazon.md>), [amazon-sagemaker](<https://devfeed.tech/tags/amazon-sagemaker.md>), [amazon-sagemaker-ai](<https://devfeed.tech/tags/amazon-sagemaker-ai.md>), [announcements](<https://devfeed.tech/tags/announcements.md>), [apis](<https://devfeed.tech/tags/apis.md>), [auto-label](<https://devfeed.tech/tags/auto-label.md>), [build](<https://devfeed.tech/tags/build.md>), [data](<https://devfeed.tech/tags/data.md>), [data-preparation](<https://devfeed.tech/tags/data-preparation.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [development](<https://devfeed.tech/tags/development.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [examples](<https://devfeed.tech/tags/examples.md>), [foundational-100](<https://devfeed.tech/tags/foundational-100.md>), [model](<https://devfeed.tech/tags/model.md>), [model-development](<https://devfeed.tech/tags/model-development.md>), [quality](<https://devfeed.tech/tags/quality.md>), [workflows](<https://devfeed.tech/tags/workflows.md>)

### AI overview

Amazon SageMaker AI adds modular data preparation capabilities that let teams generate synthetic examples, label datasets, review uncertain labels, assess quality, and transform data before training or evaluation. Auto-labeling supports built-in and custom tasks; for tasks with fixed answers, multiple models label each row and their agreement supplies a confidence score, with low-confidence rows routed to human review in SageMaker Studio. Synthetic data can be grounded in seed data and documents, and generated pipelines can be inspected and modified. Teams can chain these steps, for example generating task-specific data, reshaping it for SFT, DPO, or RL, and checking for duplicates, leakage, and rule violations. Work is driven through Python SDK APIs or an AI-ML agent skill, with artifacts open to inspection and editing. Data stays in the team's S3 buckets in its chosen Region under an IAM execution role it controls. The capabilities are generally available in US East (N. Virginia) and US West (Oregon).

### Source excerpt

Amazon SageMaker AI now offers AI data preparation capabilities for building high-quality training and evaluation data. Generate synthetic examples, assess and improve datasets, auto-label at scale, and bring in expert review through Python SDK APIs and an agent skill, all from your own notebook or IDE.

## Claude as a Teammate: A Practical Guide to AI-Powered Marketing Workflows

DevFeed: [Claude as a Teammate: A Practical Guide to AI-Powered Marketing Workflows](<https://devfeed.tech/articles/claude-as-a-teammate-a-practical-guide-to-ai-powered-marketing-workflows-138688.md>)

Original publisher: [Read original article](<https://hackernoon.com/claude-as-a-teammate-a-practical-guide-to-ai-powered-marketing-workflows?source=rss>)

Author: Daryna

Published: 2026-11-17T17:22:33Z

Content type: tutorial

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [Claude](<https://devfeed.tech/topics/claude.md>), [Claude Code](<https://devfeed.tech/topics/claude-code.md>), [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [MSP MCP](<https://devfeed.tech/topics/msp-mcp.md>), [Search engine optimization (SEO)](<https://devfeed.tech/topics/seo.md>), [Website](<https://devfeed.tech/topics/website.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>), [Slack](<https://devfeed.tech/topics/slack.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-for-marketing](<https://devfeed.tech/tags/ai-for-marketing.md>), [ai-in-marketing](<https://devfeed.tech/tags/ai-in-marketing.md>), [articles](<https://devfeed.tech/tags/articles.md>), [b2b](<https://devfeed.tech/tags/b2b.md>), [case](<https://devfeed.tech/tags/case.md>), [case-studies](<https://devfeed.tech/tags/case-studies.md>), [claude](<https://devfeed.tech/tags/claude.md>), [claude-cowork](<https://devfeed.tech/tags/claude-cowork.md>), [connectors](<https://devfeed.tech/tags/connectors.md>), [cowork](<https://devfeed.tech/tags/cowork.md>), [engineers](<https://devfeed.tech/tags/engineers.md>), [enterprise-ai](<https://devfeed.tech/tags/enterprise-ai.md>), [generative-ai](<https://devfeed.tech/tags/generative-ai.md>), [good-company](<https://devfeed.tech/tags/good-company.md>), [guide](<https://devfeed.tech/tags/guide.md>), [marketing](<https://devfeed.tech/tags/marketing.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [pages](<https://devfeed.tech/tags/pages.md>), [seo](<https://devfeed.tech/tags/seo.md>), [skills](<https://devfeed.tech/tags/skills.md>), [team](<https://devfeed.tech/tags/team.md>), [uses](<https://devfeed.tech/tags/uses.md>), [waiting](<https://devfeed.tech/tags/waiting.md>), [workflows](<https://devfeed.tech/tags/workflows.md>)

### AI overview

BotsCrew describes using Claude as a shared marketing teammate connected to company systems, with shared instructions and repeatable playbooks. Five workflows turn records from Jira, Confluence and Fathom into reviewed case studies; use Claude Design and Claude Code to build website pages; use Ahrefs via MCP to research AI-search articles; create a Gmail domain-highlighting extension; and schedule a morning digest from email and Slack. The article reports case studies shrinking from 1-2 weeks to about 2 hours, custom pages from weeks to 2-3 hours, blog posts from 2-3 hours to about 30 minutes, and link-status checks from 1-2 hours daily to zero. It emphasizes checking outputs and sources, human approval before publishing or sending, and technical review of website code and deployment. It says the approach fits recurring internal work with accessible data, and is unsuitable as a substitute for building a customer-facing product.

### Source excerpt

How a B2B marketing team uses Claude Cowork, skills and connectors to ship case studies, web pages and SEO articles without waiting on engineers.

## 106 Agent Skills That Write Copy, Price Products and Run Your Go-to-Market

DevFeed: [106 Agent Skills That Write Copy, Price Products and Run Your Go-to-Market](<https://devfeed.tech/articles/106-agent-skills-that-write-copy-price-products-and-run-your-go-to-market-138028.md>)

Original publisher: [Read original article](<https://hackernoon.com/106-agent-skills-that-write-copy-price-products-and-run-your-go-to-market?source=rss>)

Author: Hack Marketing with HackerNoon for Businesses

Published: 2026-11-17T15:15:02Z

Content type: release

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-skills](<https://devfeed.tech/tags/agent-skills.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [command](<https://devfeed.tech/tags/command.md>), [copy](<https://devfeed.tech/tags/copy.md>), [copywriting](<https://devfeed.tech/tags/copywriting.md>), [copywriting-skills](<https://devfeed.tech/tags/copywriting-skills.md>), [email](<https://devfeed.tech/tags/email.md>), [gives](<https://devfeed.tech/tags/gives.md>), [go-to-market](<https://devfeed.tech/tags/go-to-market.md>), [go-to-market-skills](<https://devfeed.tech/tags/go-to-market-skills.md>), [gtm-agent-skills](<https://devfeed.tech/tags/gtm-agent-skills.md>), [launches](<https://devfeed.tech/tags/launches.md>), [marketing](<https://devfeed.tech/tags/marketing.md>), [one](<https://devfeed.tech/tags/one.md>), [open](<https://devfeed.tech/tags/open.md>), [products](<https://devfeed.tech/tags/products.md>), [run](<https://devfeed.tech/tags/run.md>), [seo](<https://devfeed.tech/tags/seo.md>), [skills](<https://devfeed.tech/tags/skills.md>), [skills-that-write-copy](<https://devfeed.tech/tags/skills-that-write-copy.md>), [write](<https://devfeed.tech/tags/write.md>)

### AI overview

One command gives an AI agent access to 106 open-licensed marketing skills spanning copywriting, SEO, pricing, cold email, launches, analytics and retention.

### Source excerpt

One command now gives an AI agent 106 open-licensed marketing skills. They cover copywriting, SEO, pricing, cold email, launches, analytics and retention.

[Next page](<https://devfeed.tech/topics/agent-skills.md?cursor=WyIyMDI2LTExLTE3VDE1OjE1OjAyKzAwOjAwIiwgIjk4MWVkYWUyLThmNzEtNGZiNy04ZDgwLTE2NGE3YjlmZDY4YyJd>)