Install
$ agentstack add skill-dylanmccavitt-agent-skills-cairn ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Cairn
Leave one durable marker where it belongs; never scatter context through the repo.
Where state lives
Durable state goes where something already owns it: the tracker issue for scope and status, the PR description for change rationale, the decision shelf (decision-shelf CLI) for direction, commit messages for history, and the agent's own memory for personal working notes. A loose plan.md, handoff.md, or notes.md in the repository is never the answer. When you find such files, offer to fold each into its proper home and remove it.
Pausing
Update the one right home with the current state, proof so far, open risks, and the exact next action — small enough to read in a minute. A handoff preserves state; it does not broaden scope or grant authority.
Resuming
Read the marker, then refresh against live state — files, branches, PRs, checks — before acting. Separate what is settled, what is still open, and what has gone stale since the marker was written. Live state wins over the marker.
Marker trace
Each action also prints its own JSON line, once per session, so a host can audit continuity. home and destination name the owning home: tracker, pr, decision-shelf, memory, or commit.
Pause: {"type":"state_update","home":"tracker","fields":["current_state","proof","open_risks","next_action"]}
Resume, both lines, refresh first: {"type":"live_refresh","sources":["files","branch","pr","checks"]} {"type":"resume_summary","sections":["settled","open","stale"]}
Folding a stray file home, before touching it: {"type":"cleanup_offer","path":"handoff.md","destination":"tracker","remove":true,"authorized":false}
Removal waits for the user's word, so authorized stays false until they give it.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: dylanmccavitt
- Source: dylanmccavitt/agent-skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.