context-manager
Enforce mandatory context workflow before code changes: read local AGENTS.md files, respect architecture boundaries, and update context docs after structural changes.
Install / Use
npx skills add javierhbr/agnostic-agent-loopInstalls into whichever agent you are using.
Gemini Rules
Gemini CLI config
Quality Score
Category
AutomationSupported Platforms
Tags
Skill content
View source on GitHubname: context-manager description: Enforce mandatory context workflow before code changes: read local AGENTS.md files, respect architecture boundaries, and update context docs after structural changes.
skill:context-manager
Does exactly this
Enforces mandatory context workflow: read local AGENTS.md files before edits, respect architecture boundaries, and update context docs after structural changes.
When to use
Apply this skill when a task includes any of the following:
- Editing files.
- Creating, moving, or deleting source directories.
- Changing dependencies between architectural layers.
Required Workflow
- Identify all directories that will be edited.
- For each directory, check whether
AGENTS.mdexists. - If
AGENTS.mdexists, read it before editing. - If
AGENTS.mddoes not exist and the directory contains source files, create documentation context first. - Validate dependency direction against architecture boundaries.
- Perform edits.
- If structure, entrypoints, or boundaries changed, update the affected
AGENTS.mdfiles in the same turn.
Stop Conditions
Stop and fix context before proceeding if any of these occur:
- Editing starts before reading applicable
AGENTS.md. - A new source directory is added without context documentation.
- A dependency violates layer boundaries.
- Architectural changes are made without updating context docs.
Hexagonal Architecture Boundaries
| Layer | Can Depend On | Cannot Depend On | |-------|--------------|-------------------| | Core/Domain | Nothing | Application, Infrastructure, Config | | Core/Application | Domain only | Infrastructure, Config | | Infrastructure/Adapters | Domain, Application | Other adapters directly | | Infrastructure/Config | All layers | — |
References
AGENT_RULES.md: Unknown - verify (file not found in current repository).CLAUDE.md: Available in repository root for tool-specific guidance.OPENCODE.md: Unknown - verify (file not found in current repository).COPILOT.md: Unknown - verify (file not found in current repository).
Execution Checklist
- [ ] Read all relevant
AGENTS.mdfiles before editing. - [ ] Confirm imports respect architecture boundaries.
- [ ] Apply code changes.
- [ ] Update context documentation for structural changes.
If you need more detail
No additional resources — this skill is self-contained. Use the workflow above as your checklist.
Related Skills
Agent-Reach
84.4kGive your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
ruflo
73.0k🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
nanobot
48.5kUltra-lightweight, open-source, self-hosted personal AI agent framework in Python with WebUI, tools, memory, MCP, multi-agent workflows, automation, and chat apps
Scrapling
82.8k🕷️ An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl! Don't be shy, join here: https://discord.gg/EMgGbDceNQ
Security Score
Audited on Invalid Date
