all-source
AGENTS.md4 starsChanged yesterdayApache-2.0View on GitHub
This project supports cn (chronis) and bd (beads) for issue tracking. Chronis is the primary tracker. Run cn init to get started, or use bd onboard for legacy beads. Planned work runs as one ritual-full flow: Think, Spec, Plan, Build, Review, Ship, Reflect. It runs on a Chronis epic whose ritual-* lines name each step's file. Flow: .claude/skills/ritual-full/SKILL.md. Guide: https://github.com/decebal/decebal-claude-skills/blob/main/docs/ritual-zero-to-hero.md. When ending a work session, you MUST complete ALL steps below. Work is NOT complete until git push succeeds. Branch…
- Commits and pushes
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 32,671 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository25 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- allsource-incident-statusSkill
- allsource-partnershipsSkill
- ritual-fullSkill
- ritualSkill
- semantic-isaSkill
- admin-healthSkill
- allsource-data-accessSkill
- allsource-debugSkill
- allsource-incident-statusSkill
- allsource-mcp-setupSkill
- allsource-onboardSkill
- chronos-data-flow-embeddedSkill
- chronos-data-flowSkill
- chronos-durabilitySkill
- chronos-embedded-durabilitySkill
- chronos-releaseSkill
- fly-statusSkill
- mammoth-memorySkill
- ritual-fullSkill
- ritualSkill
- semantic-isaSkill
- seo-aeo-auditSkill
- allsource-customerSkill
- allsource-prime-memorySkill
The file
Show its rules and text
We split this AGENTS.md into the rules it gives an agent
15 rules. About 0 are common, 9 are distinctive, and 6 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈9
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- could be automatic: a setting could enforce it
- Distinctive: line 43
File issues for remaining work: - Create issues for anything that needs follow-up
- Distinctive: line 55
Verify: - All changes committed AND pushed to main
- Distinctive: line 60
NEVER stop before pushing - that leaves work stranded locally
- Distinctive: line 61
NEVER say "ready to push when you are" - YOU must push (except for releases — tag pushes still need explicit user confirmation per the …
- Distinctive: line 62
NEVER force-push main. If you need to rewrite history, stop and ask.
- Distinctive: line 63
If push fails, resolve and retry until it succeeds
- Distinctive: line 64
Always verify git branch --show-current returns main before committing — there is a known branch-slip quirk where HEAD can flip between …
- Very specific: line 105
Start: Run cn ready to find actionable work
- Very specific: line 109
Sync: Always run cn sync --git at session end
- Very specific: line 116
Parent-child: Use --parent=<id> to create hierarchy under epics
- Very specific: line 133
Check cn ready at session start to find available work
- Distinctive: line 134
Update status as you work (claim → done)
- Very specific: line 135
Create new tasks with cn task create when you discover work
- Distinctive: line 136
Use descriptive titles and set appropriate priority/type
- Very specific: line 137
Always cn sync --git before ending session
≈ “Common” is an estimate: at least 1 in 5 of the 41,581 AGENTS.md 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 Instructions
This project supports cn (chronis) and bd (beads) for issue tracking. Chronis is the primary tracker. Run cn init to get started, or use bd onboard for legacy beads.
Quick Reference (Chronis — preferred)
cn ready # Find available work (open + unblocked)
cn show <id> # View task details, children, timeline
cn claim <id> # Claim a task (uses CN_AGENT_ID env var)
cn done <id> # Complete a task
cn sync --git # Sync with git (add .chronis/, commit, push)Quick Reference (Beads — legacy)
bd ready # Find available work
bd show <id> # View issue details
bd update <id> --status in_progress # Claim work
bd close <id> # Complete work
bd sync # Sync with gitPlanning
Planned work runs as one ritual-full flow: Think, Spec, Plan, Build, Review, Ship, Reflect. It runs on a Chronis epic whose ritual-* lines name each step's file. Flow: .claude/skills/ritual-full/SKILL.md. Guide: https://github.com/decebal/decebal-claude-skills/blob/main/docs/ritual-zero-to-hero.md.
- Ship follows the Git Workflow in
CLAUDE.md. It opens a PR only when the user asked for one; otherwise it commits and pushes tomain. - Design approval is the user's decision. Approve only after the user tells you to approve in the conversation, and quote that instruction in your report. Never approve on your own initiative.
- Approve with
cos ritual approve <epic> <design> --by "<name>" --review <path> --yes, orcn approve <epic>withoutcos.--yesneeds decebal/chief-of-staff-v2#23. - A bug with a clear reproduction skips Think and Spec and goes through
systematic-debugging. .claude/skills/ritual*andsemantic-isaare vendored. Do not edit them here; change them upstream in decebal/decebal-claude-skills.
Landing the Plane (Session Completion)
When ending a work session, you MUST complete ALL steps below. Work is NOT complete until git push succeeds.
Branch policy: This repo uses main-branch-first — commit and push directly to
main, no feature branches, no PRs for routine work. See the Git Workflow section ofCLAUDE.mdfor the full rules, exceptions, and the release-tag carve-out. The workflow below assumesmainis your working branch.
MANDATORY WORKFLOW:
- File issues for remaining work - Create issues for anything that needs follow-up
- Run quality gates (if code changed) - Tests, linters, builds
- Update issue status - Close finished work, update in-progress items
- PUSH TO REMOTE - This is MANDATORY:
git branch --show-current # MUST print "main" — verify before pushing git pull --rebase cn sync --git # or: bd sync git push origin main git status # MUST show "up to date with origin/main" - Clean up - Clear stashes, prune stale local branches
- Verify - All changes committed AND pushed to
main - Hand off - Provide context for next session
CRITICAL RULES:
- Work is NOT complete until
git pushsucceeds - NEVER stop before pushing - that leaves work stranded locally
- NEVER say "ready to push when you are" - YOU must push (except for releases — tag pushes still need explicit user confirmation per the release skill)
- NEVER force-push
main. If you need to rewrite history, stop and ask. - If push fails, resolve and retry until it succeeds
- Always verify
git branch --show-currentreturnsmainbefore committing — there is a known branch-slip quirk where HEAD can flip between tool calls
<!-- bv-agent-instructions-v1 -->Chronis Workflow Integration
This project uses chronis as the primary event-sourced task tracker. Tasks are stored in .chronis/ and tracked in git.
Essential Commands
# Task management
cn task create "Title" -p p0 --type=epic # Create epic
cn task create "Subtask" --parent=<id> -p p1 # Create child task
cn task create "Bug fix" --type=bug # Create bug
cn list # All tasks
cn list --status=open # Filter by status
cn ready # Unblocked open tasks
cn show <id> # Details + children + timeline
cn claim <id> # Claim (reads CN_AGENT_ID)
cn done <id> --reason="Completed" # Complete
cn approve <id> # Approve
# Dependencies
cn dep add <task-id> <blocker-id> # Add blocker
cn dep remove <task-id> <blocker-id> # Remove blocker
# Migration & sync
cn migrate-beads # Import from .beads/
cn sync --git # Git add/commit/push .chronis/
# Visualization
cn tui # Interactive TUI
cn serve # Web viewer (port 3905)Workflow Pattern
- Start: Run
cn readyto find actionable work - Claim: Use
cn claim <id> - Work: Implement the task
- Complete: Use
cn done <id> - Sync: Always run
cn sync --gitat session end
Key Concepts
- Dependencies: Tasks can block other tasks.
cn readyshows only unblocked work. - Priority: p0=critical, p1=high, p2=medium, p3=low
- Types: task, epic, bug, feature
- Parent-child: Use
--parent=<id>to create hierarchy under epics - Blocking:
cn dep add <task> <blocker>to add post-creation dependencies
Session Protocol
Before ending any session, run this checklist:
git status # Check what changed
git add <files> # Stage code changes
cn sync --git # Commit chronis changes
git commit -m "..." # Commit code
git push # Push to remoteBest Practices
- Check
cn readyat session start to find available work - Update status as you work (claim → done)
- Create new tasks with
cn task createwhen you discover work - Use descriptive titles and set appropriate priority/type
- Always
cn sync --gitbefore ending session
Beads Workflow Integration (Legacy)
This project also has beads_viewer for issue tracking. Issues are stored in .beads/ and tracked in git.
Essential Commands
# View issues (launches TUI - avoid in automated sessions)
bv
# CLI commands for agents (use these instead)
bd ready # Show issues ready to work (no blockers)
bd list --status=open # All open issues
bd show <id> # Full issue details with dependencies
bd create --title="..." --type=task --priority=2
bd update <id> --status=in_progress
bd close <id> --reason="Completed"
bd close <id1> <id2> # Close multiple issues at once
bd sync # Commit and push changesKey Concepts
- Dependencies: Issues can block other issues.
bd readyshows only unblocked work. - Priority: P0=critical, P1=high, P2=medium, P3=low, P4=backlog (use numbers, not words)
- Types: task, bug, feature, epic, question, docs
- Blocking:
bd dep add <issue> <depends-on>to add dependencies
<!-- end-bv-agent-instructions -->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.

