Creates a durable work checkpoint by committing progress and publishing a resume note for cold-start agents to resume tasks.
- GitHub stars
- 113
- Files
- 3
- Last commit
- Nov 1, 2026
- License
- MIT
- Allowed tools
- —
What does the make-camp skill do?
make-camp creates a durable work checkpoint by committing progress and publishing a resume note for cold-start agents to resume tasks.
Summary generated from the skill's SKILL.md.
About the make-camp skill
Developers managing long-running AI tasks can use this skill to create durable work checkpoints before an agent goes offline or undergoes context compaction. It allows an agent to stop at a safe boundary by committing uncommitted edits as a single wip commit and writing a detailed resume note. This note captures intent, progress, current state, and next steps to ensure a cold-start agent can pick up exactly where the previous session left off.
To use this, an agent must follow a strict protocol of stopping at an atomic step and avoiding irreversible actions like new pull requests. The process relies on the scripts/resume.mjs tool to publish the checkpoint, which verifies links and files before updating the project pointer. By using the checkpoint directory defined in the resume-storage.md reference, the skill ensures that the state survives interruptions. It is designed specifically for explicit pauses rather than continuous work, producing a structured handoff that includes key files and any user-requested artifacts.
What an agent can do with it
- 1Saving progress before a session restartAn agent can commit its current work and write a resume note so it can continue the task after a restart.
- 2Managing context compaction limitsWhen an agent's context window is nearly full, it can use this skill to archive its current state and next steps.
- 3Preserving user-requested artifacts during a pauseThe skill can save specific questions or checklists that a user wants to remain available for the next session.
Written from the skill's SKILL.md. Every use case is checked against a quote from it.
How to install the make-camp skill
Each command copies the skill folder into the client's skills directory. Restart the client or start a new session to load it.Install the make-camp skill in Plugin
/plugin marketplace add tmcfarlane/team-avatar /plugin install team-avatar@team-avatar
Install the make-camp skill in Claude Code
~/.claude/skills/make-camp/SKILL.md
$ npx degit tmcfarlane/team-avatar/plugins/team-avatar/skills/make-camp ~/.claude/skills/make-campInstall the make-camp skill in Claude Code (project)
.claude/skills/make-camp/SKILL.md
$ npx degit tmcfarlane/team-avatar/plugins/team-avatar/skills/make-camp .claude/skills/make-campInstall the make-camp skill in Codex
~/.agents/skills/make-camp/SKILL.md
$ npx degit tmcfarlane/team-avatar/plugins/team-avatar/skills/make-camp ~/.agents/skills/make-campInstall the make-camp skill in Gemini CLI
~/.gemini/skills/make-camp/SKILL.md
$ npx degit tmcfarlane/team-avatar/plugins/team-avatar/skills/make-camp ~/.gemini/skills/make-campInstall the make-camp skill in Claude.ai
npx degit tmcfarlane/team-avatar/plugins/team-avatar/skills/make-camp make-camp && zip -r make-camp.zip make-camp # claude.ai → Customize → Skills → + → Create skill → Upload a skill → make-camp.zip
Automated checks
Ran Nov 14, 2026Full report →Passed our automated checks. That is not a security audit: read the scripts before you let an agent run them. How checks work
- Source repository linked and reachabletmcfarlane/team-avatar/plugins/team-avatar/skills/make-camp is public and reachablegithub.com/tmcfarlane/team-avatar/tree/HEAD/plugins/team-ava ↗Pass
- Open source license (OSI)MITspdx.org/licenses/MIT.html ↗Pass
- Commit or release in the last yearLast commit 2026-11-01github.com/tmcfarlane/team-avatar/commits ↗Pass
- No download-and-run or credential theft in scripts1 script read; no download-and-run, remote eval, base64 execution or credential uploadPass
- No hidden instructions in SKILL.mdNo override phrases, safeguard switches, credential requests or hidden charactersgithub.com/tmcfarlane/team-avatar/blob/HEAD/plugins/team-ava ↗Pass
- Pre-approved tools are scopedNo pre-approved tools; the agent asks before using onegithub.com/tmcfarlane/team-avatar/blob/HEAD/plugins/team-ava ↗Pass
- SKILL.md follows the Agent Skills formatValid name and description; says when to use it; 34 linesgithub.com/tmcfarlane/team-avatar/blob/HEAD/plugins/team-ava ↗Pass
SKILL.md
View on GitHubAppa's camp
When the team sets down for the night, they make camp. The fire is banked, the saddle is packed, and the map is out where anyone can read it. Whoever wakes first knows exactly where the trip left off. You own a clean stop. Leave a checkpoint a cold-start agent can resume from with break-camp.
On Codex, read avatar-mode/references/codex.md for tool equivalents.
This runs only on an explicit pause. On "keep going", "going to bed, keep going", or "don't stop", do not pause.
Steps
-
Stop at a safe boundary. Finish the current atomic step or back out of it. Start nothing new, and stop any characters you dispatched.
-
Take no irreversible action to pause. No new PR and no push unless you already had one out.
-
Make the work durable. Commit uncommitted edits as one clear
wip:commit on the current branch, so nothing is lost. If the tree is broken, say so in one line of the commit body. -
Write the resume note outside your context. Capture the intent, what you were doing, progress and what is verified, the current state, the next steps, the key files, and the gotchas. Use the checkpoint directory from references/resume-storage.md, which defines the shared writer and reader locator for every runtime. It must survive the expected interruption. If a
battle-mapdecision trail exists, point at it instead of duplicating it. -
Preserve what the user asked to keep. Keep the artifacts the user explicitly requested for use after resume, such as questions, an acceptance checklist, or a procedure. Reuse the existing artifact when there is one. Otherwise save the requested material beside the resume note and link it. Preserve exact wording, completeness, and order when requested. This does not mean archiving every aside or pausing an ongoing task.
-
Publish and read it back. Publish the checkpoint through scripts/resume.mjs, which verifies links and files before it atomically updates the project pointer.
node <this-skill>/scripts/resume.mjs begin --project <worktree> node <this-skill>/scripts/resume.mjs publish --project <worktree> --note <dir>/resume.md --artifact <file> ... node <this-skill>/scripts/resume.mjs read --project <worktree>Read it using only the project directory, and check the requested contents before you declare the checkpoint ready. Ten requested questions means all ten are reachable from the note alone.
Reply
Where you are in the loop. What is on disk versus still only in your head (paths, no diff dumps). The commits you made and whether the tree is clean. The first action on resume. This is a pause, not a final report.
Files in this skill
3make-camp skill questions
How do I install the make-camp skill in Claude Code?
Run npx degit tmcfarlane/team-avatar/plugins/team-avatar/skills/make-camp ~/.claude/skills/make-camp. Claude Code loads skills from ~/.claude/skills (all projects) or .claude/skills (one project) and uses a skill when its description matches the task.
Which clients can use the make-camp skill?
Any client that reads the Agent Skills format: Claude Code, Codex, Gemini CLI and claude.ai (upload as a ZIP). Scripts in the skill run only where the client allows code execution.
When should I use the make-camp skill?
Use it for explicit pauses such as 'pause here', 'stop for now', 'save your place', session restarts, or when imminent context compaction is expected.
How does the skill handle uncommitted work?
The skill commits uncommitted edits as a single clear 'wip:' commit on the current branch to ensure no work is lost.
What information should be included in the resume note?
The note should capture intent, progress, verified status, current state, next steps, key files, and any potential gotchas.