agent-init
Skill11 starsChanged 18 days agoMITView on GitHub
Analyze a workspace and generate or maintain a modular project documentation setup with a lean, concise agents.md entrypoint. Governs user directives, precautions, and desired persona, while providing selective on-demand reading of detailed guide files and immediate on-the-fly updates when code or rules change. Triggers: 'agent init', 'agents.md 생성', '워크스페이스 분석', 'agent guideline 생성', 'agents.md 모듈화', 'agent-init'. Not for LibrAgent runtime/harness facts (use libragent-harness-reference).
2 other projects use a version of this.
Adopted by
Most recently taken up by
- rgb-24bit/agent-skillstheir own version, first seen Mar 23, 2026
- dyxushuai/agent-skillstheir own version, first seen Jan 27, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- dyxushuai/agent-skills★ 2same name
- rgb-24bit/agent-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository48 other files it gives its agents.
- agents.mdAGENTS.md
- .agents/skills/react-best-practices/AGENTS.mdAGENTS.md
- .agents/skills/react-native-skills/AGENTS.mdAGENTS.md
- .agents/skills/react-view-transitions/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- agy-delegateSkill
- audit-docsSkill
- code-audit-expertSkill
- computer-diagnosisSkill
- create-builtin-toolSkill
- create-prSkill
- critique-builtin-toolSkill
- doc-auditorSkill
- docxSkill
- extract-log-debugSkill
- feature-refactor-proposalSkill
- gatekeeper-loopSkill
- harbor-harness-improvement-loopSkill
- i18n-language-standardsSkill
- jobs-trace-analyzerSkill
- lean-builtin-tool-auditorSkill
- libragent-qualitySkill
- libr-delegateSkill
- libr-tool-e2e-harness-loopSkill
- modular-large-file-refactorSkill
- performance-auditSkill
- pptxSkill
- vercel-react-best-practicesSkill
- vercel-react-native-skillsSkill
- vercel-react-view-transitionsSkill
- readme-showcase-syncSkill
- refactor-builtin-toolSkill
- release-managerSkill
- review-local-changesSkill
- review-pr-commentsSkill
- science-searchSkill
- setup-wizardSkill
- skill-creatorSkill
- soul-awakeningSkill
- sqlite-analyzerSkill
- sync-pending-approvalsSkill
- tool-creatorSkill
- trace-analyzerSkill
- trace-eval-loopSkill
- vitepress-doc-syncSkill
- worker-loopSkill
- x-cliSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
13 rules. About 0 are common, 4 are distinctive, and 9 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
9
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 30
When an agent discovers changed file paths, refactored modules, new build scripts, or updated user instructions, it immediately updates the …
- Very specific: line 42
Build & Language Configs: Check package.json, Cargo.toml, pyproject.toml, go.mod, etc.
- Very specific: line 49
Ensure docs/guidelines/ exists and generate the modular files using templates from references/templates/:
- Very specific: line 65
Create or update agents.md at the workspace root using references/templates/agents-md-template.md:
- Distinctive: line 67
Keep agents.md concise (<200 lines).
- Very specific: line 69
Ensure all pointer links to docs/guidelines/*.md are valid relative paths.
- Distinctive: line 73
When executing future agent tasks in this workspace:
- Distinctive: line 77
Read agents.md for overall routing.
- Very specific: line 78
Read docs/guidelines/architecture-and-files.md before refactoring or locating code.
- Very specific: line 79
Read docs/guidelines/coding-standards.md before writing/testing code.
- Very specific: line 80
Read docs/guidelines/persona-and-rules.md when verifying tone, user preferences, or compliance.
- Very specific: line 84
If code structure, file paths, commands, or user directives change during work, immediately update the target guide file in …
- Very specific: line 96
Update Protocol Guide: references/update-protocol.md
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Agent Init: Modular Guidelines & Persona Governance
This skill analyzes a workspace and establishes or updates a modular documentation structure with a lean agents.md entrypoint.
It manages user directives, strict precautions, and agent persona while keeping agents.md lightweight to minimize system prompt context bloat.
For how LibrAgent loads agents.md / SOUL.md, tool naming, and session isolation, use libragent-harness-reference instead of baking that into agents.md.
💡 Core Principles
- Lean Entrypoint (
agents.md):agents.mdserves as a quick-reference index and command cheat sheet (<200 lines).- Contains high-level project summary, critical user directives, desired persona, cheat sheet commands, and pointers to detailed guide files.
- Modular Guide Architecture (
docs/guidelines/):- Detailed specifications are split into dedicated files inside
docs/guidelines/:architecture-and-files.md: System layout, key file map, module boundaries.coding-standards.md: Style, type safety, error handling, logging, testing, validation commands.persona-and-rules.md: User instructions, desired persona/vibe, strict prohibitions, user preferences.workflows.md: Dev workflows, CI/CD pipelines, release procedures.
- Detailed specifications are split into dedicated files inside
- Selective Reading (Progressive Disclosure):
- Agents read
agents.mdon startup and selectively load relevantdocs/guidelines/*.mdfiles on demand based on current task scope.
- Agents read
- Self-Healing & Immediate Updates:
- When an agent discovers changed file paths, refactored modules, new build scripts, or updated user instructions, it immediately updates the relevant guide file and syncs
agents.md.
- When an agent discovers changed file paths, refactored modules, new build scripts, or updated user instructions, it immediately updates the relevant guide file and syncs
- Governance of User Directives & Persona:
- User instructions, prohibitions, and desired persona are preserved across sessions and strictly governed.
📋 Step-by-Step Execution Workflow
Step 1: Explore Workspace & Discover Information
Scan the workspace root to collect facts:
- Build & Language Configs: Check
package.json,Cargo.toml,pyproject.toml,go.mod, etc. - Directory Structure: Inspect depth-2 tree (
src/,src-tauri/,docs/,lib/,tests/). - Commands: Extract scripts for dev, build, test, lint, and validation.
- User Directives & Persona: Extract existing user instructions, persona files (
SOUL.md), or explicit chat instructions.
Step 2: Create Modular Guide Files in docs/guidelines/
Ensure docs/guidelines/ exists and generate the modular files using templates from references/templates/:
docs/guidelines/persona-and-rules.md:- Fill using
references/templates/persona-and-rules-template.md. - Record desired agent persona/vibe, user directives, strict prohibitions, and language preferences.
- Fill using
docs/guidelines/architecture-and-files.md:- Fill using
references/templates/architecture-template.md. - Record system architecture diagram, key directory map, key file locations, and layering rules.
- Fill using
docs/guidelines/coding-standards.md:- Fill using
references/templates/coding-standards-template.md. - Record formatting/linting rules, type safety rules, error handling conventions, and validation commands.
- Fill using
docs/guidelines/workflows.md:- Record dev workflow steps, CI/CD pipeline triggers, and PR requirements.
Step 3: Generate or Update Lean agents.md Entrypoint
Create or update agents.md at the workspace root using references/templates/agents-md-template.md:
- Keep
agents.mdconcise (<200 lines). - Fill project overview, high-level persona & user directives, command cheat sheet, and guide file index.
- Ensure all pointer links to
docs/guidelines/*.mdare valid relative paths.
Step 4: Apply On-Demand Reading & Immediate Update Protocols
When executing future agent tasks in this workspace:
📖 Selective Reading Protocol
- Read
agents.mdfor overall routing. - Read
docs/guidelines/architecture-and-files.mdbefore refactoring or locating code. - Read
docs/guidelines/coding-standards.mdbefore writing/testing code. - Read
docs/guidelines/persona-and-rules.mdwhen verifying tone, user preferences, or compliance.
🔄 Immediate Update Protocol (Self-Healing)
- If code structure, file paths, commands, or user directives change during work, immediately update the target guide file in
docs/guidelines/usingworkspace__strReplace. - Sync
agents.mdif high-level cheat sheets or summaries changed. - See
references/update-protocol.mdfor details.
🛠️ Resources & Templates
- Entrypoint Template:
references/templates/agents-md-template.md - Persona & Rules Template:
references/templates/persona-and-rules-template.md - Architecture Template:
references/templates/architecture-template.md - Coding Standards Template:
references/templates/coding-standards-template.md - Update Protocol Guide:
references/update-protocol.md - Merge Policy Guide:
references/merge-policy.md
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

