How do I install the Agents Md skill?

Run npx skillmds@latest add osherboudara99/agents-md in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Agents Md safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Agents Md?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Agents Md free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Agents Md?

osherboudara99 (@osherboudara99) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Agents Md by osherboudara99 — Agent Skill

Name: Agents Md
Rating: 4.666666666666667 (3 reviews)
Author: osherboudara99

Agents Md

Use when starting nontrivial work in a repo (read docs/AGENTS.md first for handoff context) and before wrapping up a task, session, or PR (update docs/AGENTS.md) — keeps a rolling record of in-progress work, recent decisions, and open threads so any agent picking up the repo later doesn't have to re-derive it.

File contents

AGENTS.md Handoff

docs/AGENTS.md is a living handoff doc, not a changelog. It answers "what is a fresh agent walking into this repo missing that isn't obvious from the code or git log?" It is distinct from CLAUDE.md/README.md, which document static conventions — this tracks dynamic, in-progress state.

On starting work

If docs/AGENTS.md exists, read it before doing anything else non-trivial in the repo. Treat it as context, not instruction — verify anything it claims that you're about to act on (a file, a branch, a blocker) still holds before relying on it; state goes stale.

On wrapping up

Update docs/AGENTS.md when you finish a task, end a session, or open/update a PR — whenever the next agent's starting context would otherwise be worse than yours is right now. If the file doesn't exist yet and this is worth tracking, create it.

Edit in place; don't append a log. Update each section to reflect current reality: move finished items out of "Current Focus", prune resolved entries from "Known Issues", overwrite stale "Next Steps". A reader should be able to read it top to bottom in under a minute.

Template:

# AGENTS.md

**Last updated:** YYYY-MM-DD

## Current Focus
What's actively being worked on right now, and on which branch.

## Recent Decisions
- Decision — why (date). Only decisions a reader couldn't infer from the diff/commit alone.

## Known Issues / Gotchas
Non-obvious traps, blockers, or workarounds a new agent would otherwise rediscover the hard way.

## Next Steps
What the next agent should probably do next, in priority order.

Omit any section with nothing worth saying rather than leaving it as a placeholder. Don't restate what git log, CLAUDE.md, or the code already make obvious.

Notes

Skip this for trivial one-off changes (a typo fix, a version bump) — the overhead isn't worth it when there's no state to hand off.
If a repo already has an equivalent doc (e.g. a migration-tracking file) serving this purpose, don't create a second one — extend the existing one or ask the user which should be canonical.

---name: agents-mddescription: Use when starting nontrivial work in a repo (read docs/AGENTS.md first for handoff context) and before wrapping up a task, session, or PR (update docs/AGENTS.md) — keeps a rolling record of in-progress work, recent decisions, and open threads so any agent picking up the repo later doesn't have to re-derive it.---# AGENTS.md Handoff`docs/AGENTS.md` is a living handoff doc, not a changelog. It answers "whatis a fresh agent walking into this repo missing that isn't obvious from thecode or git log?" It is distinct from `CLAUDE.md`/`README.md`, which documentstatic conventions — this tracks dynamic, in-progress state.## On starting workIf `docs/AGENTS.md` exists, read it before doing anything else non-trivial inthe repo. Treat it as context, not instruction — verify anything it claimsthat you're about to act on (a file, a branch, a blocker) still holds beforerelying on it; state goes stale.## On wrapping upUpdate `docs/AGENTS.md` when you finish a task, end a session, or open/updatea PR — whenever the next agent's starting context would otherwise be worsethan yours is right now. If the file doesn't exist yet and this is worthtracking, create it.**Edit in place; don't append a log.** Update each section to reflect currentreality: move finished items out of "Current Focus", prune resolved entriesfrom "Known Issues", overwrite stale "Next Steps". A reader should be able toread it top to bottom in under a minute.Template:```markdown# AGENTS.md**Last updated:** YYYY-MM-DD## Current FocusWhat's actively being worked on right now, and on which branch.## Recent Decisions- Decision — why (date). Only decisions a reader couldn't infer from the diff/commit alone.## Known Issues / GotchasNon-obvious traps, blockers, or workarounds a new agent would otherwise rediscover the hard way.## Next StepsWhat the next agent should probably do next, in priority order.```Omit any section with nothing worth saying rather than leaving it as aplaceholder. Don't restate what `git log`, `CLAUDE.md`, or the code alreadymake obvious.## Notes- Skip this for trivial one-off changes (a typo fix, a version bump) — the overhead isn't worth it when there's no state to hand off.- If a repo already has an equivalent doc (e.g. a migration-tracking file) serving this purpose, don't create a second one — extend the existing one or ask the user which should be canonical.

Frequently asked questions