# Swamp Club Feed

Writing shared by Swamp Club operatives. Judged by the Gator; only the real stuff makes it here.

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

## Automate the Status Update, Not the Conversation

DevFeed: [Automate the Status Update, Not the Conversation](<https://devfeed.tech/articles/automate-the-status-update-not-the-conversation-146578.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/fa87e965-c1d9-4363-b828-a3a51d09a425/go>)

Author: jpisgeek

Published: 2026-11-20T14:57:52Z

Content type: article

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [AI for data automation](<https://devfeed.tech/topics/ai-for-data-automation.md>), [agent to agent communication](<https://devfeed.tech/topics/agent-to-agent-communication.md>), [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>), [rachels-ramblings](<https://devfeed.tech/topics/rachels-ramblings.md>)

Tags: [agent-workflows](<https://devfeed.tech/tags/agent-workflows.md>), [agents](<https://devfeed.tech/tags/agents.md>), [automate](<https://devfeed.tech/tags/automate.md>), [azure](<https://devfeed.tech/tags/azure.md>), [azure-devops](<https://devfeed.tech/tags/azure-devops.md>), [clickup](<https://devfeed.tech/tags/clickup.md>), [dashboard](<https://devfeed.tech/tags/dashboard.md>), [dashboards](<https://devfeed.tech/tags/dashboards.md>), [discord](<https://devfeed.tech/tags/discord.md>), [docs](<https://devfeed.tech/tags/docs.md>), [drop-in](<https://devfeed.tech/tags/drop-in.md>), [files](<https://devfeed.tech/tags/files.md>), [fleet](<https://devfeed.tech/tags/fleet.md>), [github](<https://devfeed.tech/tags/github.md>), [know](<https://devfeed.tech/tags/know.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>), [status](<https://devfeed.tech/tags/status.md>), [talk](<https://devfeed.tech/tags/talk.md>), [time](<https://devfeed.tech/tags/time.md>), [update](<https://devfeed.tech/tags/update.md>)

### AI overview

The author wants agents to make project catch-up self-service so team conversations can focus on difficult decisions. Existing agents update multiple tools, but coworkers still have to check them and piece together changes; shared notes or a docs repository help preserve and exchange information, yet still require syncing and review. The proposed starting point is one project and one question: can a teammate learn what changed since yesterday without asking the author to reconcile several sources? Answers should identify what changed and when, link to supporting tickets or runs, and disclose stale information or uncertainty. The examples show why: a VM build reported "WORKGROUP joined" despite a successful domain join, and a dashboard that reports only the last healthy check-in can conceal a later failed attempt. The team approach remains unproven; the author suggests seeing whether better-prepared calls make room for substantive discussion.

### Source excerpt

What if agents handled the catch-up, and we used that time to talk about the hard stuff?

## Scored Confidence Agent Workflows in the Decision-Making Process

DevFeed: [Scored Confidence Agent Workflows in the Decision-Making Process](<https://devfeed.tech/articles/scored-confidence-agent-workflows-in-the-decision-making-process-134694.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/4e1fbb30-0889-49fb-80a5-0f9c17518481/go>)

Author: cjd

Published: 2026-11-15T18:55:24Z

Content type: article

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [text-classification](<https://devfeed.tech/topics/text-classification.md>), [decision-making](<https://devfeed.tech/topics/decision-making.md>), [human review](<https://devfeed.tech/topics/human-review.md>), [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [AI Integration Strategies](<https://devfeed.tech/topics/ai-integration-strategies.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Tool](<https://devfeed.tech/topics/tool.md>), [LLM security](<https://devfeed.tech/topics/llm-security.md>)

Tags: [accountability](<https://devfeed.tech/tags/accountability.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agent-workflows](<https://devfeed.tech/tags/agent-workflows.md>), [agents](<https://devfeed.tech/tags/agents.md>), [classification](<https://devfeed.tech/tags/classification.md>), [confidence](<https://devfeed.tech/tags/confidence.md>), [decision](<https://devfeed.tech/tags/decision.md>), [decision-making](<https://devfeed.tech/tags/decision-making.md>), [executive](<https://devfeed.tech/tags/executive.md>), [humans](<https://devfeed.tech/tags/humans.md>), [in-the-loop](<https://devfeed.tech/tags/in-the-loop.md>), [language](<https://devfeed.tech/tags/language.md>), [language-model](<https://devfeed.tech/tags/language-model.md>), [making](<https://devfeed.tech/tags/making.md>), [model](<https://devfeed.tech/tags/model.md>), [models](<https://devfeed.tech/tags/models.md>), [process](<https://devfeed.tech/tags/process.md>), [swamp](<https://devfeed.tech/tags/swamp.md>), [tasks](<https://devfeed.tech/tags/tasks.md>), [workflows](<https://devfeed.tech/tags/workflows.md>)

### AI overview

The workflow treats suitable language model decisions as classification and leaves high-stakes executive decisions to human review. Deterministic scripts gather accessible facts, while an LLM handles language judgments; its proposed classifications are scored by a separate deterministic system against explicit criteria, then gated before presentation with supporting evidence and criteria met or unmet. Using email sorting as an example, the article lists sender, subject, body, and attachment signals, including numeric and yes-or-no facts. It describes Swamp as the pipeline engine for fetching and parsing data, invoking an agent, structuring its output, applying criteria, and presenting results, with scheduled or triggered runs and built-in encryption for secrets. The author says this approach can use any language model, including Claude, while acknowledging unresolved ethical concerns and the models' unpredictability.

### Source excerpt

A developer describes using scored confidence workflows with any language model via Swamp Club's Swamp framework to handle decision-making as classification, keeping humans in the loop for high-stakes executive decisions while delegating lower-level classification tasks to models.

## Governance in the Age of Agents

DevFeed: [Governance in the Age of Agents](<https://devfeed.tech/articles/governance-in-the-age-of-agents-129415.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/aadaf249-22ea-440e-9850-f292b4dc03d3/go>)

Author: stack72

Published: 2026-11-11T13:34:17Z

Content type: opinion

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [ai-governance](<https://devfeed.tech/topics/ai-governance.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [human review](<https://devfeed.tech/topics/human-review.md>), [qualitygates](<https://devfeed.tech/topics/qualitygates.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>)

Tags: [adversarial](<https://devfeed.tech/tags/adversarial.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [attestation](<https://devfeed.tech/tags/attestation.md>), [audit](<https://devfeed.tech/tags/audit.md>), [blast-radius](<https://devfeed.tech/tags/blast-radius.md>), [bug-fix](<https://devfeed.tech/tags/bug-fix.md>), [building-with-ai](<https://devfeed.tech/tags/building-with-ai.md>), [context](<https://devfeed.tech/tags/context.md>), [contracts](<https://devfeed.tech/tags/contracts.md>), [decision](<https://devfeed.tech/tags/decision.md>), [decision-making](<https://devfeed.tech/tags/decision-making.md>), [development](<https://devfeed.tech/tags/development.md>), [governance](<https://devfeed.tech/tags/governance.md>), [production](<https://devfeed.tech/tags/production.md>), [ship](<https://devfeed.tech/tags/ship.md>), [tests](<https://devfeed.tech/tags/tests.md>), [verification](<https://devfeed.tech/tags/verification.md>)

### AI overview

Agent governance should surface consequential assumptions before they become code and keep a record of how work was verified and changed. The author illustrates the risk with an agent's reasonable manifest-path plan that conflicted with an undocumented byte-for-byte contract; human review caught it before implementation. At Swamp, work moves through recorded lifecycle states: an adversarial plan check and human approval precede implementation; independent verification runs against the exact commit; failed attempts are logged; and conformance review compares the result with the approved plan, requiring explanations for deviations before a PR can open. The author says scrutiny should scale with ambiguity, blast radius, reversibility, and product or architectural judgment, while keeping records of attempts and verification. These are practices built for Swamp's work, not presented as a universal template.

### Source excerpt

An agent can turn an unexamined assumption into code, tests, and docs that all agree with each other and ship it in one session. Governance is how you make the decisions, verification, and deviations visible before that assumption reaches production.

## I Asked Swamp to Turn My Linux VM Process into a Windows One - A Delve into the Realm of Technology and the Human Condition

DevFeed: [I Asked Swamp to Turn My Linux VM Process into a Windows One - A Delve into the Realm of Technology and the Human Condition](<https://devfeed.tech/articles/i-asked-swamp-to-turn-my-linux-vm-process-into-a-windows-one-a-delve-into-the-realm-of-technology-and-the-human-condition-123600.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/df2c7e4a-b63b-46fd-b883-ba97882dcfef/go>)

Author: jpisgeek

Published: 2026-11-09T11:51:26Z

Content type: article

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [Provisioning](<https://devfeed.tech/topics/provisioning.md>), [configuration-management](<https://devfeed.tech/topics/configuration-management.md>), [human review](<https://devfeed.tech/topics/human-review.md>), [workflow automation](<https://devfeed.tech/topics/workflow-automation.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [approval-gates](<https://devfeed.tech/tags/approval-gates.md>), [boot](<https://devfeed.tech/tags/boot.md>), [branch](<https://devfeed.tech/tags/branch.md>), [bug](<https://devfeed.tech/tags/bug.md>), [build](<https://devfeed.tech/tags/build.md>), [cloud-init](<https://devfeed.tech/tags/cloud-init.md>), [cluster](<https://devfeed.tech/tags/cluster.md>), [configuration](<https://devfeed.tech/tags/configuration.md>), [conversion](<https://devfeed.tech/tags/conversion.md>), [copy-and-paste](<https://devfeed.tech/tags/copy-and-paste.md>), [dns](<https://devfeed.tech/tags/dns.md>), [dynamic-dns](<https://devfeed.tech/tags/dynamic-dns.md>), [golden-image](<https://devfeed.tech/tags/golden-image.md>), [hyper-v](<https://devfeed.tech/tags/hyper-v.md>), [linux](<https://devfeed.tech/tags/linux.md>), [process](<https://devfeed.tech/tags/process.md>), [swamp](<https://devfeed.tech/tags/swamp.md>), [vm](<https://devfeed.tech/tags/vm.md>), [windows](<https://devfeed.tech/tags/windows.md>)

### AI overview

Starting from a tested Linux VM build process, the author used Swamp to adapt it for Windows VMs in about 90 seconds. The Windows workflow replaced cloud images and cloud-init with a sysprepped golden image, an answer file, and steps such as disk expansion and domain joining. Swamp held the VM roster and rendered scripts, but could not reach hosts in another datacenter, so the author transferred files manually and verified them with hashes and counts before running preflight, build, status, and configuration checks. Testing exposed copy errors, stale files, host-placement issues, and misleading domain-join reporting; the author added DNS and activation checks, preflight and what-if options, and clearer results. NetBox changes were staged in a branch, reviewed and merged manually, then checked in the main data; a merge without the Commit checkbox was only a dry run. Direct execution by a Swamp worker remains proposed.

### Source excerpt

A Delve into the Realm of Technology and the Human Condition. Writing about technology, experience, and human judgment.

## My Agent Tagged the Whole SDR Team on a Thursday Night

DevFeed: [My Agent Tagged the Whole SDR Team on a Thursday Night](<https://devfeed.tech/articles/my-agent-tagged-the-whole-sdr-team-on-a-thursday-night-119014.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/ee878065-03b9-4ea8-97cc-5d6c6902cd9f/go>)

Author: tituscapilnean

Published: 2026-11-05T05:54:44Z

Content type: opinion

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>), [Claude Code](<https://devfeed.tech/topics/claude-code.md>), [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [batch](<https://devfeed.tech/tags/batch.md>), [checkpoint](<https://devfeed.tech/tags/checkpoint.md>), [claude](<https://devfeed.tech/tags/claude.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [code](<https://devfeed.tech/tags/code.md>), [confidence](<https://devfeed.tech/tags/confidence.md>), [control](<https://devfeed.tech/tags/control.md>), [database](<https://devfeed.tech/tags/database.md>), [dry](<https://devfeed.tech/tags/dry.md>), [errors](<https://devfeed.tech/tags/errors.md>), [execution](<https://devfeed.tech/tags/execution.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [incident](<https://devfeed.tech/tags/incident.md>), [incidents](<https://devfeed.tech/tags/incidents.md>), [learn](<https://devfeed.tech/tags/learn.md>), [ledger](<https://devfeed.tech/tags/ledger.md>), [list](<https://devfeed.tech/tags/list.md>), [sdr](<https://devfeed.tech/tags/sdr.md>), [team](<https://devfeed.tech/tags/team.md>)

### AI overview

A Claude Code workflow matched campaign responders to ICP accounts and assigned SDRs, then posted a large Slack list before review. It tagged the whole team and included non-ICP accounts because rules learned from earlier, smaller runs had normalized posting and failed to make the ICP filter non-negotiable. The author deleted the post and added three safeguards: explicit approval before posting, Salesforce ICP status as the strict eligibility rule, and short summaries for large batches. The added approval slows the process, but the author considers that worthwhile to protect SDR attention and trust in the leads.

### Source excerpt

I've missed sharing stories on this blog. My last post here was back in May, and it's time to get back into it.

## Testing Blender Scenes Built by LLM Agents

DevFeed: [Testing Blender Scenes Built by LLM Agents](<https://devfeed.tech/articles/testing-blender-scenes-built-by-llm-agents-111705.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/8bf3b9da-054e-4d96-9969-f2a9ff976377/go>)

Author: aaddrick

Published: 2026-11-01T18:49:13Z

Content type: article

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [blender](<https://devfeed.tech/topics/blender.md>), [3D](<https://devfeed.tech/topics/3d.md>), [Test automation](<https://devfeed.tech/topics/test-automation.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [Claude Code](<https://devfeed.tech/topics/claude-code.md>)

Tags: [3d](<https://devfeed.tech/tags/3d.md>), [acceleration](<https://devfeed.tech/tags/acceleration.md>), [actuator](<https://devfeed.tech/tags/actuator.md>), [add-ons](<https://devfeed.tech/tags/add-ons.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-disclosure](<https://devfeed.tech/tags/ai-disclosure.md>), [alignment](<https://devfeed.tech/tags/alignment.md>), [blender](<https://devfeed.tech/tags/blender.md>), [claude](<https://devfeed.tech/tags/claude.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [clipping](<https://devfeed.tech/tags/clipping.md>), [film](<https://devfeed.tech/tags/film.md>), [llm](<https://devfeed.tech/tags/llm.md>), [llm-agents](<https://devfeed.tech/tags/llm-agents.md>), [motion](<https://devfeed.tech/tags/motion.md>), [scale](<https://devfeed.tech/tags/scale.md>), [testing](<https://devfeed.tech/tags/testing.md>), [viewport](<https://devfeed.tech/tags/viewport.md>)

### AI overview

LLM agents building Blender scenes through headless bpy need scripted checks against the evaluated scene because they cannot inspect the viewport and can mistake their build instructions for verified results. The article organizes checks into 12 families, from mesh integrity, specification and rig structure to collisions, contact, motion, camera visibility, renders, pipeline isolation and visual review. It recommends read-only verification by a separate agent, measurements in world space across relevant frames, mutation tests that prove each checker catches planted defects, and results tied to the exact .blend file and spec version. Collision checks need care: nearest-face normals can under-report penetration in merged meshes, while ray parity can over-report on overlapping shells; the project required both methods to agree. The advice is qualified by project-specific tolerances, and the author notes that foot-plant and interpenetration checks were not run on the shipped file and that its render bypassed an outdated default spec.

### Source excerpt

An agent building a Blender scene can't see the viewport, so every claim it makes about clipping, contact, scale or motion has to be measured. These are the twelve families of checks from a 640-frame mech film a team of Claude agents built in bpy, plus the three times the checks themselves were...

## Replacing the newsroom with one workflow | 💻Gerard Klijs🦀

DevFeed: [Replacing the newsroom with one workflow | 💻Gerard Klijs🦀](<https://devfeed.tech/articles/replacing-the-newsroom-with-one-workflow-gerard-klijs-109210.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/b16a82d3-c672-4d88-85c8-6664757726e7/go>)

Author: gklijs

Published: 2026-10-30T07:39:31Z

Content type: article

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [workflow automation](<https://devfeed.tech/topics/workflow-automation.md>), [Orchestration](<https://devfeed.tech/topics/orchestration.md>), [Claude](<https://devfeed.tech/topics/claude.md>), [Image](<https://devfeed.tech/topics/image.md>), [Claude Code](<https://devfeed.tech/topics/claude-code.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [conference](<https://devfeed.tech/tags/conference.md>), [diffusion](<https://devfeed.tech/tags/diffusion.md>), [digest](<https://devfeed.tech/tags/digest.md>), [event](<https://devfeed.tech/tags/event.md>), [event-sourcing](<https://devfeed.tech/tags/event-sourcing.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [illustration](<https://devfeed.tech/tags/illustration.md>), [local-model](<https://devfeed.tech/tags/local-model.md>), [memory](<https://devfeed.tech/tags/memory.md>), [minutes](<https://devfeed.tech/tags/minutes.md>), [model](<https://devfeed.tech/tags/model.md>), [news](<https://devfeed.tech/tags/news.md>), [out-of-memory](<https://devfeed.tech/tags/out-of-memory.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [repo](<https://devfeed.tech/tags/repo.md>), [swamp](<https://devfeed.tech/tags/swamp.md>), [video](<https://devfeed.tech/tags/video.md>), [workflow](<https://devfeed.tech/tags/workflow.md>)

### AI overview

The author describes a one-person newsroom workflow that runs twice each weekday: one run covers neurodiversity and the other basic income. Each takes a couple of minutes to gather news and social posts, summarize them with a local model, email a digest, draft a LinkedIn post, and generate an illustration with Stable Diffusion XL. Both topics use the same workflow. The author switched from qwen3.6:35b-a3b, which ran out of memory, to the lighter qwen3:14b. They attribute the workflow's event-sourcing approach to keeping a record of each run, with changes added as new layers rather than silently replacing prior state. The excerpt does not provide further technical details.

### Source excerpt

Yesterday at the AxonIQ conference I learned about brag, a Claude Code skill that turns a repository into a short launch video. So today I pointed it at my own repo, and it made a 20-second pitch deck out of a swamp workflow. The workflow is my one-person newsroom. Every weekday afternoon it runs...

## I forked my own software, seven years late - reset.dev

DevFeed: [I forked my own software, seven years late - reset.dev](<https://devfeed.tech/articles/i-forked-my-own-software-seven-years-late-reset-dev-102116.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/5281fd55-57bd-48f9-8271-9309ce8017f6/go>)

Author: reset

Published: 2026-10-26T17:18:51Z

Content type: article

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [experience report](<https://devfeed.tech/topics/experience-report.md>), [Operating system](<https://devfeed.tech/topics/operating-system.md>), [free software](<https://devfeed.tech/topics/free-software.md>), [Tool](<https://devfeed.tech/topics/tool.md>), [Bash](<https://devfeed.tech/topics/bash.md>), [virtual machines](<https://devfeed.tech/topics/virtual-machines.md>), [versioning](<https://devfeed.tech/topics/versioning.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>)

Tags: [agent-harness](<https://devfeed.tech/tags/agent-harness.md>), [ai-coding-agents](<https://devfeed.tech/tags/ai-coding-agents.md>), [apache-2-0](<https://devfeed.tech/tags/apache-2-0.md>), [apple](<https://devfeed.tech/tags/apple.md>), [arm64](<https://devfeed.tech/tags/arm64.md>), [automation](<https://devfeed.tech/tags/automation.md>), [bash](<https://devfeed.tech/tags/bash.md>), [bugs](<https://devfeed.tech/tags/bugs.md>), [build](<https://devfeed.tech/tags/build.md>), [chef](<https://devfeed.tech/tags/chef.md>), [cli](<https://devfeed.tech/tags/cli.md>), [dependencies](<https://devfeed.tech/tags/dependencies.md>), [dojo](<https://devfeed.tech/tags/dojo.md>), [extension](<https://devfeed.tech/tags/extension.md>), [habitat](<https://devfeed.tech/tags/habitat.md>), [operating-system](<https://devfeed.tech/tags/operating-system.md>), [software](<https://devfeed.tech/tags/software.md>), [swamp](<https://devfeed.tech/tags/swamp.md>)

### AI overview

Dojo forks Habitat at its last release without a license prompt (0.79.1), retaining Apache 2.0 and the original project history. The fork is being developed into an operating system with a Launcher as PID 1, package-based machine installation, and Supervisor gossip rings. An agent harness helped repair the test setup and translate 625 package plans from Bash to Fennel, using a translator, differential tests, and regeneration checks. The Swamp extension orders plan dependencies into 88 build waves and records builds; it handles the 43-plan base-toolchain cycle with a manually ordered base list and marks plans blocked when dependencies fail in the same run. Dojo currently relies on Nix for its build environment and booted system. The planned path away from that dependency requires isolated, repeatable build passes and updates to its older base toolchain.

### Source excerpt

The author describes forking Habitat (a Chef-originated software packaging and supervision tool) at its last Apache 2.0-licensed release from 2019, renaming it Dojo, and rebuilding its 625 plans from bash into Fennel while developing it into a minimal operating system.

## What the Swamp registry reads from your extension - reset.dev

DevFeed: [What the Swamp registry reads from your extension - reset.dev](<https://devfeed.tech/articles/what-the-swamp-registry-reads-from-your-extension-reset-dev-102115.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/2d6e7a07-784c-44df-ab38-2d46e55cf776/go>)

Author: reset

Published: 2026-10-26T16:29:25Z

Content type: tutorial

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [Test-driven development](<https://devfeed.tech/topics/tdd.md>), [Documentation](<https://devfeed.tech/topics/documentation.md>), [code beautification](<https://devfeed.tech/topics/code-beautification.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [argument](<https://devfeed.tech/tags/argument.md>), [arguments](<https://devfeed.tech/tags/arguments.md>), [automation](<https://devfeed.tech/tags/automation.md>), [behavior](<https://devfeed.tech/tags/behavior.md>), [catalog](<https://devfeed.tech/tags/catalog.md>), [cli](<https://devfeed.tech/tags/cli.md>), [deno](<https://devfeed.tech/tags/deno.md>), [dry-run](<https://devfeed.tech/tags/dry-run.md>), [export](<https://devfeed.tech/tags/export.md>), [extension](<https://devfeed.tech/tags/extension.md>), [extensions](<https://devfeed.tech/tags/extensions.md>), [object](<https://devfeed.tech/tags/object.md>), [regex](<https://devfeed.tech/tags/regex.md>), [registry](<https://devfeed.tech/tags/registry.md>), [rules](<https://devfeed.tech/tags/rules.md>), [runtime](<https://devfeed.tech/tags/runtime.md>), [scanning](<https://devfeed.tech/tags/scanning.md>), [source](<https://devfeed.tech/tags/source.md>), [swamp](<https://devfeed.tech/tags/swamp.md>), [text](<https://devfeed.tech/tags/text.md>), [tools](<https://devfeed.tech/tags/tools.md>)

### AI overview

The Swamp registry extracts extension arguments from source text with regular expressions; it does not inspect runtime zod schemas, so unmatched fields and descriptions silently disappear even when the quality grade is 100%. To make fields visible, write them directly inside `z.object({ ... })`, begin each field on one line as `name: z.type(`, use a plain string literal in `.describe("...")`, wrap ISO datetime validation in `z.string().pipe(...)`, and declare `version` as a string literal. The free `swamp extension push --dry-run` JSON checks global arguments only, so it cannot catch missing method arguments. The article recommends a source-reading test for all schemas, followed by reading back `swamp extension info <name> --json` after publishing to compare listed arguments with schema fields.

### Source excerpt

The Swamp registry extracts extension argument names and descriptions via regex scanning of source text rather than runtime introspection, with specific formatting rules around .describe(), line breaks, and z.object() structure determining what appears on registry pages.

## Showcase: Security Scanning with Swamp

DevFeed: [Showcase: Security Scanning with Swamp](<https://devfeed.tech/articles/showcase-security-scanning-with-swamp-102805.md>)

Original publisher: [Read original article](<https://www.youtube.com/watch?v=NuxZGEYowqA>)

Author: skunk-ape

Published: 2026-10-26T00:26:03Z

Content type: tutorial

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [Security](<https://devfeed.tech/topics/security.md>)

Tags: [camera-phone](<https://devfeed.tech/tags/camera-phone.md>), [compromise](<https://devfeed.tech/tags/compromise.md>), [emudhra](<https://devfeed.tech/tags/emudhra.md>), [emudhra-ca-compromise](<https://devfeed.tech/tags/emudhra-ca-compromise.md>), [free](<https://devfeed.tech/tags/free.md>), [php](<https://devfeed.tech/tags/php.md>), [rce](<https://devfeed.tech/tags/rce.md>), [scan](<https://devfeed.tech/tags/scan.md>), [scanning](<https://devfeed.tech/tags/scanning.md>), [security](<https://devfeed.tech/tags/security.md>), [security-scanning](<https://devfeed.tech/tags/security-scanning.md>), [sharing](<https://devfeed.tech/tags/sharing.md>), [showcase](<https://devfeed.tech/tags/showcase.md>), [swamp](<https://devfeed.tech/tags/swamp.md>), [system](<https://devfeed.tech/tags/system.md>), [upload](<https://devfeed.tech/tags/upload.md>), [using](<https://devfeed.tech/tags/using.md>), [video](<https://devfeed.tech/tags/video.md>), [video-phone](<https://devfeed.tech/tags/video-phone.md>)

### AI overview

The showcase demonstrates using Swamp to scan a system for signs of eMudhra CA compromise and PHP remote code execution (RCE).

### Source excerpt

Using Swamp to scan a system for eMudhra CA compromise and PHP RCE.

## Telling swamp what I want

DevFeed: [Telling swamp what I want](<https://devfeed.tech/articles/telling-swamp-what-i-want-98829.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/80cc98cb-c8a1-4732-9669-260272972930/go>)

Author: svendowideit

Published: 2026-10-24T09:56:58Z

Content type: article

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [static-site-generator](<https://devfeed.tech/topics/static-site-generator.md>), [Markdown](<https://devfeed.tech/topics/markdown.md>), [Obsidian](<https://devfeed.tech/topics/obsidian-md.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>)

Tags: [backing-up](<https://devfeed.tech/tags/backing-up.md>), [bash](<https://devfeed.tech/tags/bash.md>), [bash-shell](<https://devfeed.tech/tags/bash-shell.md>), [beam](<https://devfeed.tech/tags/beam.md>), [blog](<https://devfeed.tech/tags/blog.md>), [caddy](<https://devfeed.tech/tags/caddy.md>), [cloudflare](<https://devfeed.tech/tags/cloudflare.md>), [cloudflare-pages](<https://devfeed.tech/tags/cloudflare-pages.md>), [credentials](<https://devfeed.tech/tags/credentials.md>), [dns](<https://devfeed.tech/tags/dns.md>), [extension](<https://devfeed.tech/tags/extension.md>), [extensions](<https://devfeed.tech/tags/extensions.md>), [git](<https://devfeed.tech/tags/git.md>), [html](<https://devfeed.tech/tags/html.md>), [inventory](<https://devfeed.tech/tags/inventory.md>), [markdown](<https://devfeed.tech/tags/markdown.md>), [obsidian](<https://devfeed.tech/tags/obsidian.md>), [repository](<https://devfeed.tech/tags/repository.md>), [swamp](<https://devfeed.tech/tags/swamp.md>), [workflow](<https://devfeed.tech/tags/workflow.md>)

### AI overview

The author built a custom swamp workflow to publish existing Obsidian blog posts without creating a separate Git repository. It uses a new @svendowideit/markdown2html extension to progressively enhance HTML using whatever frontmatter is present, and @svendowideit/caddy to set up DNS through the author's existing libdns zone. Publishing to Cloudflare Pages requires a token with Cloudflare Pages edit access and DNS metadata zone scope; the token and account ID are stored in a swamp vault. The workflow is run with inputs for the blog and publish directories, hostname, vault, Caddy and DNS models, site title, and template path. The author says the implementation took a few hours and can later be invoked through opencode by asking it to redeploy the site.

### Source excerpt

A developer builds a custom swamp workflow that converts Obsidian markdown blog posts to HTML and publishes them to Cloudflare Pages with automatic DNS setup via Caddy, without requiring a separate git repository.

## Setting up Swamp in a C# and Unity shop - reset.dev

DevFeed: [Setting up Swamp in a C# and Unity shop - reset.dev](<https://devfeed.tech/articles/setting-up-swamp-in-a-c-and-unity-shop-reset-dev-98622.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/3a94fb08-19f3-4ca9-aca3-5d97bc0fa992/go>)

Author: reset

Published: 2026-10-23T23:14:07Z

Content type: tutorial

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [developer setup](<https://devfeed.tech/topics/developer-setup.md>), [Developer experience](<https://devfeed.tech/topics/developer-experience.md>), [workflow automation](<https://devfeed.tech/topics/workflow-automation.md>), [TypeScript](<https://devfeed.tech/topics/typescript.md>), [C#](<https://devfeed.tech/topics/csharp.md>), [Unity](<https://devfeed.tech/topics/unity.md>)

Tags: [agent-harness](<https://devfeed.tech/tags/agent-harness.md>), [ai-tool](<https://devfeed.tech/tags/ai-tool.md>), [api-key](<https://devfeed.tech/tags/api-key.md>), [audit](<https://devfeed.tech/tags/audit.md>), [auth](<https://devfeed.tech/tags/auth.md>), [automation](<https://devfeed.tech/tags/automation.md>), [behavior](<https://devfeed.tech/tags/behavior.md>), [behaviors](<https://devfeed.tech/tags/behaviors.md>), [browser](<https://devfeed.tech/tags/browser.md>), [building](<https://devfeed.tech/tags/building.md>), [c-sharp](<https://devfeed.tech/tags/c-sharp.md>), [cli](<https://devfeed.tech/tags/cli.md>), [game](<https://devfeed.tech/tags/game.md>), [setup](<https://devfeed.tech/tags/setup.md>), [swamp](<https://devfeed.tech/tags/swamp.md>), [telemetry](<https://devfeed.tech/tags/telemetry.md>), [tools](<https://devfeed.tech/tags/tools.md>), [transparency](<https://devfeed.tech/tags/transparency.md>), [typescript](<https://devfeed.tech/tags/typescript.md>), [unity](<https://devfeed.tech/tags/unity.md>), [workflow](<https://devfeed.tech/tags/workflow.md>)

### AI overview

In a C# and Unity studio, Swamp provided auditable behavior runs that retain results and record who initiated them, with workflow runs tagging records by run. Setup had practical friction: Homebrew 7 required trusting the tap, browser sign-in needed a real terminal, and workflow examples needed object-form `dependsOn` entries and an `expr` assert field; validation returned a Zod stack trace where run gave clearer feedback. Because `swamp repo init` can install skills globally and configure an agent, the author used `--tool none` and kept skills in one repository. A blocked-command telemetry check found command path, argument names, result metadata, version and environment details, but no file contents or stdout. The studio keeps Swamp in an operations repository and is building extensions for game and studio tools. These are first-day findings from the supplied excerpt.

### Source excerpt

A developer documents the first-day experience configuring Swamp, a TypeScript-based automation CLI, in a C# and Unity game studio, covering installation quirks, behavior and workflow setup, agent scoping with --tool none, and telemetry transparency.

## How Much of Your Software Factory Can Run in the Dark?

DevFeed: [How Much of Your Software Factory Can Run in the Dark?](<https://devfeed.tech/articles/how-much-of-your-software-factory-can-run-in-the-dark-96308.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/60a5f4f1-f7f0-4456-a0f8-6534f8d7bfc3/go>)

Author: stack72

Published: 2026-10-22T20:38:34Z

Content type: opinion

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [engineering-culture](<https://devfeed.tech/topics/engineering-culture.md>), [implementation](<https://devfeed.tech/topics/implementation.md>), [human review](<https://devfeed.tech/topics/human-review.md>), [Test automation](<https://devfeed.tech/topics/test-automation.md>), [human computer interaction](<https://devfeed.tech/topics/human-computer-interaction.md>), [Developer experience](<https://devfeed.tech/topics/developer-experience.md>), [Integration testing](<https://devfeed.tech/topics/integration-testing.md>)

Tags: [about](<https://devfeed.tech/tags/about.md>), [account](<https://devfeed.tech/tags/account.md>), [actions](<https://devfeed.tech/tags/actions.md>), [added](<https://devfeed.tech/tags/added.md>), [agent](<https://devfeed.tech/tags/agent.md>), [build](<https://devfeed.tech/tags/build.md>), [building-with-ai](<https://devfeed.tech/tags/building-with-ai.md>), [builds](<https://devfeed.tech/tags/builds.md>), [can](<https://devfeed.tech/tags/can.md>), [comes](<https://devfeed.tech/tags/comes.md>), [factory](<https://devfeed.tech/tags/factory.md>), [giving](<https://devfeed.tech/tags/giving.md>), [let](<https://devfeed.tech/tags/let.md>), [responsibility](<https://devfeed.tech/tags/responsibility.md>), [run](<https://devfeed.tech/tags/run.md>), [software](<https://devfeed.tech/tags/software.md>), [software-factory](<https://devfeed.tech/tags/software-factory.md>), [systems](<https://devfeed.tech/tags/systems.md>), [want](<https://devfeed.tech/tags/want.md>), [with](<https://devfeed.tech/tags/with.md>), [work](<https://devfeed.tech/tags/work.md>)

### AI overview

The article argues that operational software can run unattended when its job, authority, response boundaries and escalation path are defined, while product development often needs human involvement to shape behavior and revise assumptions. Agents can investigate and implement features, but seemingly clear tickets leave consequential choices--for example, whether workflow retries restart from the beginning or resume after failure. Discussing such choices during development helps avoid embedding unwanted behavior in code, tests and documentation. The author also recommends using agents to explore and try alternatives, not simply ship the first working version. Both kinds of work need independent verification: agent-written tests can confirm a mistaken interpretation, and automated checks cannot determine whether the intended behavior was chosen. The boundary shifts as operational incidents expose new problems or product decisions become settled. Repeated questions should be captured in context or checks; human involvement is most useful where decisions still need discussion.

### Source excerpt

Our lights-on software factory builds systems we want to operate with the lights off. We work through the responsibility we're giving those systems, build it into the operation and let them run. When something new comes back, we have more work to do in the software factory.

## Asking swamp first, even for this blog -- Notes

DevFeed: [Asking swamp first, even for this blog -- Notes](<https://devfeed.tech/articles/asking-swamp-first-even-for-this-blog-notes-95919.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/9dc467d1-ec8b-4277-9ac3-e38e55b9fb25/go>)

Author: tagur

Published: 2026-10-22T15:42:10Z

Content type: article

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [Tool](<https://devfeed.tech/topics/tool.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [Pull Request](<https://devfeed.tech/topics/pull-request.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [api-key](<https://devfeed.tech/tags/api-key.md>), [apis](<https://devfeed.tech/tags/apis.md>), [arguments](<https://devfeed.tech/tags/arguments.md>), [automation](<https://devfeed.tech/tags/automation.md>), [behaviors](<https://devfeed.tech/tags/behaviors.md>), [blog](<https://devfeed.tech/tags/blog.md>), [bugs](<https://devfeed.tech/tags/bugs.md>), [build-tools](<https://devfeed.tech/tags/build-tools.md>), [chef](<https://devfeed.tech/tags/chef.md>), [ci](<https://devfeed.tech/tags/ci.md>), [ci-cd](<https://devfeed.tech/tags/ci-cd.md>), [cloudflare](<https://devfeed.tech/tags/cloudflare.md>), [cloudflare-pages](<https://devfeed.tech/tags/cloudflare-pages.md>), [community](<https://devfeed.tech/tags/community.md>), [custom-domain](<https://devfeed.tech/tags/custom-domain.md>), [dashboard](<https://devfeed.tech/tags/dashboard.md>), [data](<https://devfeed.tech/tags/data.md>), [network](<https://devfeed.tech/tags/network.md>), [swamp](<https://devfeed.tech/tags/swamp.md>)

### AI overview

The author uses swamp as a first stop for read-only questions and built this blog with existing extensions, storing a Cloudflare token in a vault and reviewing rendered HTML before writing it. Integrations for Meraki Dashboard and PowerShell also show how typed arguments, methods, stored data, and vault secrets can connect existing APIs; the author says the second integration cost less, while the value of reuse remains uncertain. Ten issues were filed through `swamp issue`; four were fixed and closed within three hours, which encouraged reporting problems instead of working around them. The blog workflow still required manual `git init` and custom-domain setup, and the Markdown extension had a PyYAML build problem on macOS. The author has not settled how swamp should fit alongside Terraform, Chef, GitLab, CI/CD, or shared team practices. Before swamp changes firewalls or DHCP, they want human review to guard against silent incorrect state.

### Source excerpt

Why I reach for swamp first, how I built this blog with it, and what I still need to work out before using it for network changes.

## Gatekeeping in the Age of AI

DevFeed: [Gatekeeping in the Age of AI](<https://devfeed.tech/articles/gatekeeping-in-the-age-of-ai-91425.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/1270596c-fbda-449a-b6e4-6b8289f7c297/go>)

Author: stack72

Published: 2026-10-20T21:25:30Z

Content type: opinion

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [engineering-culture](<https://devfeed.tech/topics/engineering-culture.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [Test automation](<https://devfeed.tech/topics/test-automation.md>), [Test-driven development](<https://devfeed.tech/topics/tdd.md>), [engineering-leadership](<https://devfeed.tech/topics/engineering-leadership.md>), [Requirements](<https://devfeed.tech/topics/requirements.md>)

Tags: [1-4](<https://devfeed.tech/tags/1-4.md>), [about](<https://devfeed.tech/tags/about.md>), [account](<https://devfeed.tech/tags/account.md>), [adopting](<https://devfeed.tech/tags/adopting.md>), [agent](<https://devfeed.tech/tags/agent.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>), [anthropic](<https://devfeed.tech/tags/anthropic.md>), [approach](<https://devfeed.tech/tags/approach.md>), [build](<https://devfeed.tech/tags/build.md>), [building-with-ai](<https://devfeed.tech/tags/building-with-ai.md>), [change](<https://devfeed.tech/tags/change.md>), [decades](<https://devfeed.tech/tags/decades.md>), [gatekeeping](<https://devfeed.tech/tags/gatekeeping.md>), [how-we-build](<https://devfeed.tech/tags/how-we-build.md>), [it-s](<https://devfeed.tech/tags/it-s.md>), [software](<https://devfeed.tech/tags/software.md>), [solved](<https://devfeed.tech/tags/solved.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

The article argues that "we already solved this" dismissals can discourage teams from learning how AI changes software development. At Swamp, writing code with agents has required changes to planning, review, testing and shipping; simply assigning agents the typing would miss much of that shift. The author acknowledges failures--misunderstood requirements, poor code and tests changed to pass--and calls for specific scrutiny of specifications and verification. It cites Anthropic's Scion, built as repository needs changed amid a rise from about 2,500 daily merges in May to over 30,000 by September, and Zed's Delta, which keeps agent conversations, code and review together while DeltaDB tracks changes and remains compatible with Git. These approaches are presented as responses to particular pressures, not universal replacements. The author urges experienced engineers to help teams test and revise processes, while warning that dismissive debate can prolong pilots or delay experimentation.

### Source excerpt

Every shift in software has been told it's already solved. With AI, that pushback has become gatekeeping, and it risks talking us out of the biggest chance in decades to change how we build.

## I made my terminal look like a swamp (and it croaks)

DevFeed: [I made my terminal look like a swamp (and it croaks)](<https://devfeed.tech/articles/i-made-my-terminal-look-like-a-swamp-and-it-croaks-87367.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/adc4c6ac-4a5e-42ec-aa46-cff9611248c3/go>)

Author: sntxrr

Published: 2026-10-14T22:05:26Z

Content type: tutorial

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [Terminal](<https://devfeed.tech/topics/terminal.md>), [Zellij](<https://devfeed.tech/topics/zellij.md>), [User interface design](<https://devfeed.tech/topics/ui-design.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [audio](<https://devfeed.tech/tags/audio.md>), [bash](<https://devfeed.tech/tags/bash.md>), [ci](<https://devfeed.tech/tags/ci.md>), [claude](<https://devfeed.tech/tags/claude.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [config](<https://devfeed.tech/tags/config.md>), [css](<https://devfeed.tech/tags/css.md>), [curl](<https://devfeed.tech/tags/curl.md>), [dry-run](<https://devfeed.tech/tags/dry-run.md>), [ghostty](<https://devfeed.tech/tags/ghostty.md>), [github](<https://devfeed.tech/tags/github.md>), [herdr](<https://devfeed.tech/tags/herdr.md>), [installer](<https://devfeed.tech/tags/installer.md>), [swamp](<https://devfeed.tech/tags/swamp.md>), [terminal](<https://devfeed.tech/tags/terminal.md>)

### AI overview

The author built matching swamp-club-inspired themes for herdr and Ghostty, using identical palette values so the terminal panes and herdr sidebar look continuous. After comparing the site's compiled colors with its live styles, they corrected an initial green-tinted palette: the actual design uses black and neutral gray, with green reserved for glow and accents. Neon green marks focus, while a calmer green marks completed agents; the author also corrected herdr status-color assumptions after inspecting a populated sidebar. Frogger-inspired notification sounds are synthesized from scratch and play only for agents in background workspaces. Installers support `--dry-run`; the herdr installer backs up the config, replaces only its `[theme]` section, validates and reloads it, and is described as idempotent. Manual setup requires avoiding duplicate TOML tables and placing Ghostty themes in `~/.config/ghostty/themes/`. The supplied excerpt also notes options to tone down the palette.

### Source excerpt

I spend most of my day in herdr now. If you haven't run into it, it's a terminal workspace manager built for coding agents: workspaces, tabs, panes, and a sidebar that tells you which agent is working, which one is done, and which one is sitting there waiting on you.

## Seven Months with Swamp

DevFeed: [Seven Months with Swamp](<https://devfeed.tech/articles/seven-months-with-swamp-86868.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/ddde30aa-ed06-490f-b25d-62e1d7266fbb/go>)

Author: webframp

Published: 2026-10-14T18:45:54Z

Content type: opinion

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [Loop Engineering](<https://devfeed.tech/topics/loop-engineering.md>), [Test-driven development](<https://devfeed.tech/topics/tdd.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [aws](<https://devfeed.tech/tags/aws.md>), [behavior](<https://devfeed.tech/tags/behavior.md>), [cloudformation](<https://devfeed.tech/tags/cloudformation.md>), [compliance](<https://devfeed.tech/tags/compliance.md>), [compose](<https://devfeed.tech/tags/compose.md>), [config](<https://devfeed.tech/tags/config.md>), [dag](<https://devfeed.tech/tags/dag.md>), [data](<https://devfeed.tech/tags/data.md>), [datadog](<https://devfeed.tech/tags/datadog.md>), [datastores](<https://devfeed.tech/tags/datastores.md>), [detection](<https://devfeed.tech/tags/detection.md>), [devops](<https://devfeed.tech/tags/devops.md>), [dns](<https://devfeed.tech/tags/dns.md>), [dynamodb](<https://devfeed.tech/tags/dynamodb.md>), [ecs](<https://devfeed.tech/tags/ecs.md>), [eventbridge](<https://devfeed.tech/tags/eventbridge.md>), [extension](<https://devfeed.tech/tags/extension.md>), [extensions](<https://devfeed.tech/tags/extensions.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [skills](<https://devfeed.tech/tags/skills.md>), [swamp](<https://devfeed.tech/tags/swamp.md>)

### AI overview

The author's central finding is that reliable agent behavior came from putting procedures in code rather than relying on prompt-based skills. In Swamp, typed model methods run against live systems and record versioned data; workflows compose those methods into a DAG, so agents choose behaviors and inspect results while procedures remain repeatable, testable, and reviewable across model upgrades. A workflow that always pulled metrics fixed an outage investigation's missed cold-start spike. Separately written observers later supplied data for a drift detector that made no AWS API calls of its own, instead normalizing stored snapshots and diffing them against a baseline. The author reports publishing 165 extensions. Load testing reached 500 ECS Fargate workers and briefly overwhelmed telemetry ingestion; DynamoDB sync tuning exposed successive bottlenecks. These tests surfaced limits, while the author identifies shared use by more than a hundred engineers as an open scaling question.

### Source excerpt

I tried agents with skills. When they failed, I found swamp.

## Not every prompt needs an LLM 🚦

DevFeed: [Not every prompt needs an LLM 🚦](<https://devfeed.tech/articles/not-every-prompt-needs-an-llm-86869.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/e5e83326-3190-40e1-aee7-78d8106f66ec/go>)

Author: mellens

Published: 2026-10-14T18:37:49Z

Content type: article

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [Large Language Model](<https://devfeed.tech/topics/llm.md>), [prompt](<https://devfeed.tech/topics/prompt.md>), [ai-experiments](<https://devfeed.tech/topics/ai-experiments.md>), [LLM observability](<https://devfeed.tech/topics/llm-observability.md>), [LLM security](<https://devfeed.tech/topics/llm-security.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [api](<https://devfeed.tech/tags/api.md>), [apis](<https://devfeed.tech/tags/apis.md>), [aws](<https://devfeed.tech/tags/aws.md>), [claude](<https://devfeed.tech/tags/claude.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [confidence](<https://devfeed.tech/tags/confidence.md>), [costs](<https://devfeed.tech/tags/costs.md>), [design](<https://devfeed.tech/tags/design.md>), [don-t](<https://devfeed.tech/tags/don-t.md>), [error](<https://devfeed.tech/tags/error.md>), [extension](<https://devfeed.tech/tags/extension.md>), [flow](<https://devfeed.tech/tags/flow.md>), [git](<https://devfeed.tech/tags/git.md>), [github](<https://devfeed.tech/tags/github.md>), [github-issue](<https://devfeed.tech/tags/github-issue.md>), [jev](<https://devfeed.tech/tags/jev.md>), [llm](<https://devfeed.tech/tags/llm.md>), [prompt](<https://devfeed.tech/tags/prompt.md>)

### AI overview

The puddlerouter prototype checks each Claude Code prompt and routes it to a deterministic Swamp recipe, a Jev yes/no decision, or Claude. Exact matches can be answered without an LLM; uncertain Jev results (below 0.7 confidence), errors, and timeouts fall back to Claude. Recipes are read-only and run unattended only from allow-listed publishers; prompts that appear to contain secrets are not sent to Jev, and the hook passes a temporary-file path rather than raw text. Caching avoids repeat Jev calls and Swamp searches. A reported open-PR query was answered by a Swamp recipe in about three seconds without calling Claude. The hook is still in shadow mode and logs intended routes without enforcing them; the decision route has seen little use, and the author is collecting more usage data and developing the recipe catalog.

### Source excerpt

Jev-likes and Swamping ftw

## 100s of custom integrations across AV & IT | Swamp

DevFeed: [100s of custom integrations across AV & IT | Swamp](<https://devfeed.tech/articles/100s-of-custom-integrations-across-av-it-swamp-86864.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/35a0f4df-b474-4cb0-958a-b190e2c9c36e/go>)

Author: dougschaefer

Published: 2026-10-14T17:46:04Z

Content type: article

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [API Economy](<https://devfeed.tech/topics/api-economy.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [apis](<https://devfeed.tech/tags/apis.md>), [automation](<https://devfeed.tech/tags/automation.md>), [av](<https://devfeed.tech/tags/av.md>), [custom](<https://devfeed.tech/tags/custom.md>), [digital-signage](<https://devfeed.tech/tags/digital-signage.md>), [enterprise-software](<https://devfeed.tech/tags/enterprise-software.md>), [erp-systems](<https://devfeed.tech/tags/erp-systems.md>), [integrations](<https://devfeed.tech/tags/integrations.md>), [it](<https://devfeed.tech/tags/it.md>), [swamp](<https://devfeed.tech/tags/swamp.md>)

### AI overview

American Sound used Swamp to build a growing library of custom integrations for the varied AV and IT systems it supports. The team uses Swamp to inspect a client's environment, learn its custom APIs and IoT interfaces, and develop automation that manages or improves existing systems. For custom digital signage, for example, they query the platform's structure, compare it with American Sound's standards, and create an implementation plan. The company says reusable technology behaviors help it build future solutions, while Swamp's controls and visibility support production AI agents. It reports moving from no production AI work to front-line agents and delivering custom solutions faster; the excerpt gives no quantified results. Swamp integrations with finance and ERP systems are also beginning to surface information for internal accounting teams.

### Source excerpt

American Sound used Swamp to build hundreds of custom integrations across AV and IT, giving its team front-line AI agents that meet customers where they are.

## Swamp Ushers in the Second AI Frontier with Behavior-Centric Design | 💻Gerard Klijs🦀

DevFeed: [Swamp Ushers in the Second AI Frontier with Behavior-Centric Design | 💻Gerard Klijs🦀](<https://devfeed.tech/articles/swamp-ushers-in-the-second-ai-frontier-with-behavior-centric-design-gerard-klijs-86870.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/f0b0fe61-ebed-4379-903f-ba9a0aa06351/go>)

Author: gklijs

Published: 2026-10-14T17:23:31Z

Content type: article

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [workflow automation](<https://devfeed.tech/topics/workflow-automation.md>), [software-architecture](<https://devfeed.tech/topics/software-architecture.md>), [Test automation](<https://devfeed.tech/topics/test-automation.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [amplify-partners](<https://devfeed.tech/tags/amplify-partners.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [automation](<https://devfeed.tech/tags/automation.md>), [behavior](<https://devfeed.tech/tags/behavior.md>), [behaviors](<https://devfeed.tech/tags/behaviors.md>), [bugs](<https://devfeed.tech/tags/bugs.md>), [build](<https://devfeed.tech/tags/build.md>), [cel](<https://devfeed.tech/tags/cel.md>), [cli](<https://devfeed.tech/tags/cli.md>), [code](<https://devfeed.tech/tags/code.md>), [command](<https://devfeed.tech/tags/command.md>), [cost](<https://devfeed.tech/tags/cost.md>), [data](<https://devfeed.tech/tags/data.md>), [design](<https://devfeed.tech/tags/design.md>), [llm](<https://devfeed.tech/tags/llm.md>), [platform](<https://devfeed.tech/tags/platform.md>), [swamp](<https://devfeed.tech/tags/swamp.md>), [token](<https://devfeed.tech/tags/token.md>)

### AI overview

Swamp's approach uses frontier AI once to build typed, deterministic behaviors, then runs those behaviors as code, aiming to make automations reproducible and reduce token use and cost. The author reports that Swamp announced general availability with $7M in seed funding and quotes company figures of 14x fewer tokens and 6x lower cost than an equivalent LangChain agent; Xero reportedly cut marketing automation work from hours or days to minutes. In a hands-on session, the author built and published a Swamp extension for the Vados static-site generator, with eight typed methods and a 13/13 quality score. They also encountered practical hurdles, including learning CEL field names, an outdated CLI, PowerShell-to-WSL quoting, and an undocumented quality schema. The author sees those knowledge and tooling gaps as problems the platform's behavior-centric design is meant to address.

### Source excerpt

I spent today inside the Swamp 🐊 A month or two ago I stumbled onto Swamp -- an automation platform with a genuinely different idea: instead of letting an LLM decide every step of your automation (fragile, token-hungry, impossible to reproduce), you use frontier intelligence once to build typed,...

## Sacrificing the IaC Sacred Cow for my Birthday

DevFeed: [Sacrificing the IaC Sacred Cow for my Birthday](<https://devfeed.tech/articles/sacrificing-the-iac-sacred-cow-for-my-birthday-86863.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/1fafe134-42d5-479e-87af-f35bfd98b589/go>)

Author: randybias

Published: 2026-10-14T15:38:19Z

Content type: opinion

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [Frontier AI](<https://devfeed.tech/topics/frontier-ai.md>), [agentic-coding](<https://devfeed.tech/topics/agentic-coding.md>), [Claude](<https://devfeed.tech/topics/claude.md>), [internal developer platform automation](<https://devfeed.tech/topics/internal-developer-platform-automation.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [2025](<https://devfeed.tech/tags/2025.md>), [3](<https://devfeed.tech/tags/3.md>), [about](<https://devfeed.tech/tags/about.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agentic-ops](<https://devfeed.tech/tags/agentic-ops.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [birthday](<https://devfeed.tech/tags/birthday.md>), [can](<https://devfeed.tech/tags/can.md>), [candles](<https://devfeed.tech/tags/candles.md>), [future](<https://devfeed.tech/tags/future.md>), [i](<https://devfeed.tech/tags/i.md>), [iac](<https://devfeed.tech/tags/iac.md>), [launches](<https://devfeed.tech/tags/launches.md>), [look](<https://devfeed.tech/tags/look.md>), [ops](<https://devfeed.tech/tags/ops.md>), [swamp](<https://devfeed.tech/tags/swamp.md>)

### AI overview

The author argues that complex operations need deterministic, reusable workflows beyond traditional infrastructure-as-code. Swamp's "behavior centric" approach uses coding models at build time to encode intent as versioned behaviors with typed inputs and outputs, isolated credentials, and deterministic control flow; models are called at runtime only for steps needing judgment. The article contrasts this with agent-centric systems, where a model repeatedly directs each step, and reports a toy trip-planner comparison using 14 times more tokens and costing six times more per plan for the agent-centric version. The author applied Swamp workflows to GPU-cluster setup and day-two operations, and used an isolated digital twin built with nested virtualization for testing. They frame autonomic operations as a way to extend IaC into troubleshooting, provisioning, drift checks, and other ongoing tasks, while human operators oversee the system.

### Source excerpt

As Swamp Club launches and I blow out candles, we can start to look towards a new future of Agentic Ops.

## I Want More from AI Than Another Agent | matgreten.dev

DevFeed: [I Want More from AI Than Another Agent | matgreten.dev](<https://devfeed.tech/articles/i-want-more-from-ai-than-another-agent-matgreten-dev-86866.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/986de6de-bbfd-4e70-839b-f21c4202c08a/go>)

Author: mgreten

Published: 2026-10-14T15:35:34Z

Content type: opinion

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [AI Integration Strategies](<https://devfeed.tech/topics/ai-integration-strategies.md>), [Functional programming](<https://devfeed.tech/topics/functional-programming.md>), [Pull Request](<https://devfeed.tech/topics/pull-request.md>), [Test automation](<https://devfeed.tech/topics/test-automation.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [behavior](<https://devfeed.tech/tags/behavior.md>), [operations](<https://devfeed.tech/tags/operations.md>), [reasoning](<https://devfeed.tech/tags/reasoning.md>), [workflows](<https://devfeed.tech/tags/workflows.md>)

### AI overview

The author argues that reusable, behavior-centric operations reduce repeated AI orchestration: define typed methods with validated inputs and outputs, then compose them into workflows that control sequence, conditions, and failure handling. In Swamp, workflows can call models for bounded tasks such as assessing PR comments, while code handles settled mechanics such as polling, running checks, and recording results; posting replies or applying changes remains a separate action. The author says this makes procedures easier to inspect and reuse across projects, but does not eliminate API failures, stale assumptions, model errors, or human responsibility. They also describe scaling back excessive checks: code can validate required fields, while judging whether a plan makes sense remains a human decision.

### Source excerpt

A developer argues for behavior-centric design over agent-centric workflows, advocating for reusable, typed operations with defined inputs and outputs that constrain when AI reasoning is invoked rather than letting models orchestrate entire processes.

## The Model Shouldn't Be Driving

DevFeed: [The Model Shouldn't Be Driving](<https://devfeed.tech/articles/the-model-shouldn-t-be-driving-86865.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/7c81aa08-ced7-4abb-86de-230279201b16/go>)

Author: kneel

Published: 2026-10-14T15:25:55Z

Content type: opinion

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [workflow automation](<https://devfeed.tech/topics/workflow-automation.md>), [software-architecture](<https://devfeed.tech/topics/software-architecture.md>), [AI Architecture](<https://devfeed.tech/topics/ai-architecture.md>), [Prompt Engineering](<https://devfeed.tech/topics/prompt-engineering.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [api](<https://devfeed.tech/tags/api.md>), [data](<https://devfeed.tech/tags/data.md>), [devops](<https://devfeed.tech/tags/devops.md>), [model](<https://devfeed.tech/tags/model.md>), [run](<https://devfeed.tech/tags/run.md>)

### AI overview

The author contrasts agent-centric AI systems, where models choose tools and direct control flow at runtime, with a behavior-centric approach built from small, typed, validated behaviors composed into deterministic workflows. Drawing on Swamp use for Heroku inventory, scheduled server patching, build-board provisioning, and cloud-resource tracking, the author argues that defining and verifying behavior ahead of execution can make automation easier to inspect and operate. Models can help build the system, but the author emphasizes that people still need to understand and check the assumptions encoded in it.

### Source excerpt

Swamp launched publicly today. I run most of my business and my homelab on it. Here is what 'behavior-centric' actually means once you have to operate the result.

## Announcing Swamp Club

DevFeed: [Announcing Swamp Club](<https://devfeed.tech/articles/announcing-swamp-club-86867.md>)

Original publisher: [Read original article](<https://swamp-club.com/feed/p/ba7742e9-ea24-4519-a2c6-7bff3566c61f/go>)

Author: keeb

Published: 2026-10-14T15:13:15Z

Content type: release

Language: en

Sources: [Swamp Club Feed](<https://devfeed.tech/sources/swamp-club-feed.md>)

Topics: [Developer experience](<https://devfeed.tech/topics/developer-experience.md>), [GitHub](<https://devfeed.tech/topics/github.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [amplify-partners](<https://devfeed.tech/tags/amplify-partners.md>), [angel-investors](<https://devfeed.tech/tags/angel-investors.md>), [bug](<https://devfeed.tech/tags/bug.md>), [business](<https://devfeed.tech/tags/business.md>), [community](<https://devfeed.tech/tags/community.md>), [company](<https://devfeed.tech/tags/company.md>), [customer-success](<https://devfeed.tech/tags/customer-success.md>), [customers](<https://devfeed.tech/tags/customers.md>), [ecosystem](<https://devfeed.tech/tags/ecosystem.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [extensions](<https://devfeed.tech/tags/extensions.md>), [feature](<https://devfeed.tech/tags/feature.md>), [finance](<https://devfeed.tech/tags/finance.md>), [forecasting](<https://devfeed.tech/tags/forecasting.md>), [network](<https://devfeed.tech/tags/network.md>), [reverse-engineering](<https://devfeed.tech/tags/reverse-engineering.md>), [software](<https://devfeed.tech/tags/software.md>), [swamp](<https://devfeed.tech/tags/swamp.md>), [tools](<https://devfeed.tech/tags/tools.md>)

### AI overview

Swamp says it is building software for people across business roles while using AI throughout its own operations, from finance and forecasting to engineering. The company reports that Swamp is used across its business, with bug reports typically reviewed in about three hours and fixed an hour later. At launch, it announced a $7 million seed round led by Amplify Partners, with participation from Encoded Ventures and angel investors. The article also reports thousands of users, more than 340 million builds, an extensions ecosystem, a writing-based community, and customers addressing business problems.

### Source excerpt

It feels weird starting a software company when everyone around you says software is dead. And look around you - just this week people are reverse engineering tools we all grew up with, every company

[Next page](<https://devfeed.tech/sources/swamp-club-feed.md?cursor=WyIyMDI2LTEwLTE0VDE1OjEzOjE1KzAwOjAwIiwgIjU5MTQyNjc4LTZmZGYtNDM3NC1hMWVmLTRjMTllZWZlNzE3NCJd>)