# semantic code navigation

Published articles for semantic code navigation.

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

## Why AI coding agents need codebase context and constraints

DevFeed: [Why AI coding agents need codebase context and constraints](<https://devfeed.tech/articles/your-ai-agent-doesn-t-know-your-codebase-s-context-or-constraints-and-that-s-costing-you-50238.md>)

Original publisher: [Read original article](<https://www.sonarsource.com/blog/your-ai-agent-doesnt-know-your-codebases-context-or-constraints/>)

Author: Prasenjit Sarkar

Published: 2026-09-14T13:00:00Z

Content type: article

Language: en

Sources: [Sonar Blog](<https://devfeed.tech/sources/sonar-blog.md>)

Topics: [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [software-architecture](<https://devfeed.tech/topics/software-architecture.md>), [Maintainability](<https://devfeed.tech/topics/maintainability.md>), [Security](<https://devfeed.tech/topics/security.md>), [Vulnerabilities](<https://devfeed.tech/topics/vulnerabilities.md>)

Tags: [agent-centric-development-cycle](<https://devfeed.tech/tags/agent-centric-development-cycle.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-code-review](<https://devfeed.tech/tags/ai-code-review.md>), [ai-coding-agents](<https://devfeed.tech/tags/ai-coding-agents.md>), [ai-generated-code](<https://devfeed.tech/tags/ai-generated-code.md>), [architectural-drift](<https://devfeed.tech/tags/architectural-drift.md>), [code-duplication](<https://devfeed.tech/tags/code-duplication.md>), [code-quality](<https://devfeed.tech/tags/code-quality.md>), [codebase-context](<https://devfeed.tech/tags/codebase-context.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [coding-personalities-of-llms](<https://devfeed.tech/tags/coding-personalities-of-llms.md>), [context-augmentation](<https://devfeed.tech/tags/context-augmentation.md>), [developer-productivity](<https://devfeed.tech/tags/developer-productivity.md>), [false-positive-rate](<https://devfeed.tech/tags/false-positive-rate.md>), [gitar](<https://devfeed.tech/tags/gitar.md>), [maintainability](<https://devfeed.tech/tags/maintainability.md>), [security](<https://devfeed.tech/tags/security.md>), [semantic-code-navigation](<https://devfeed.tech/tags/semantic-code-navigation.md>), [software-architecture](<https://devfeed.tech/tags/software-architecture.md>), [sonar-vortex](<https://devfeed.tech/tags/sonar-vortex.md>), [sonarqube](<https://devfeed.tech/tags/sonarqube.md>), [sonarqube-cloud](<https://devfeed.tech/tags/sonarqube-cloud.md>), [static-analysis](<https://devfeed.tech/tags/static-analysis.md>), [technical-debt](<https://devfeed.tech/tags/technical-debt.md>), [token-usage](<https://devfeed.tech/tags/token-usage.md>)

### AI overview

This article explains how AI coding agents can introduce architectural drift, maintainability problems, and security risks when they lack knowledge of a codebase's architecture, conventions, and constraints. It argues that teams should guide agents with codebase context before relying on verification and review.

### Source excerpt

Learn why AI coding agents need codebase context to reduce architectural drift, security risks, review costs, and unnecessary token usage.

## AGENTS.md is a workaround, not a solution

DevFeed: [AGENTS.md is a workaround, not a solution](<https://devfeed.tech/articles/agents-md-is-a-workaround-not-a-solution-50220.md>)

Original publisher: [Read original article](<https://www.sonarsource.com/blog/agents-md-is-a-workaround-not-a-solution/>)

Author: Prasenjit Sarkar

Published: 2026-09-14T13:00:00Z

Content type: opinion

Language: en

Sources: [Sonar Blog](<https://devfeed.tech/sources/sonar-blog.md>)

Topics: [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [Documentation](<https://devfeed.tech/topics/documentation.md>), [software-architecture](<https://devfeed.tech/topics/software-architecture.md>), [Security](<https://devfeed.tech/topics/security.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>)

Tags: [agent-centric-development-cycle](<https://devfeed.tech/tags/agent-centric-development-cycle.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agents-md](<https://devfeed.tech/tags/agents-md.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent-instructions](<https://devfeed.tech/tags/ai-agent-instructions.md>), [ai-coding-agents](<https://devfeed.tech/tags/ai-coding-agents.md>), [ai-generated-code](<https://devfeed.tech/tags/ai-generated-code.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [architecture-guidance](<https://devfeed.tech/tags/architecture-guidance.md>), [code-intelligence](<https://devfeed.tech/tags/code-intelligence.md>), [code-smells](<https://devfeed.tech/tags/code-smells.md>), [coding-standards](<https://devfeed.tech/tags/coding-standards.md>), [context-engineering](<https://devfeed.tech/tags/context-engineering.md>), [context-rot](<https://devfeed.tech/tags/context-rot.md>), [developer-guidelines](<https://devfeed.tech/tags/developer-guidelines.md>), [governance](<https://devfeed.tech/tags/governance.md>), [llm-instruction-following](<https://devfeed.tech/tags/llm-instruction-following.md>), [long-context-llms](<https://devfeed.tech/tags/long-context-llms.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [prompt-files](<https://devfeed.tech/tags/prompt-files.md>), [security](<https://devfeed.tech/tags/security.md>), [semantic-code-navigation](<https://devfeed.tech/tags/semantic-code-navigation.md>), [sonarqube](<https://devfeed.tech/tags/sonarqube.md>), [sonarqube-cloud](<https://devfeed.tech/tags/sonarqube-cloud.md>), [static-analysis](<https://devfeed.tech/tags/static-analysis.md>), [technical-debt](<https://devfeed.tech/tags/technical-debt.md>)

### AI overview

This opinion argues that AGENTS.md is useful for stable project instructions such as dependency installation, testing, development-server startup, and pull-request requirements, but is not sufficient as governance for evolving architecture, coding standards, security posture, or reuse decisions. Because it is static Markdown that depends on human maintenance, it can become outdated as the codebase changes.

### Source excerpt

AGENTS.md has won the fight to become the standard place to brief a coding agent, and deservedly so.