save
Skill7 starsChanged 10 days agoMITView on GitHub
Checkpoint work: commit, push, update the PR, wait for CI, return the preview, record session facts. Never implements or merges.
36 other projects use a version of this.
Adopted by
Most recently taken up by
- FabianoArthur/claude-code-kitchentheir own version, first seen Nov 8, 2026
- paulogabriel/beryltheir own version, first seen Nov 8, 2026
- pche-broken-artist/forge-of-thoughttheir own version, first seen Nov 6, 2026
- egregore-labs/egregoretheir own version, first seen Nov 6, 2026
- vidit19sharma/neogentstheir own version, first seen Nov 1, 2026
- 0xUrsanomics/utopia-ostheir own version, first seen Nov 1, 2026
Used with
Projects with it also use load, ingest and 9 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
TheLatentSpaceLabSkillload
Load a document into memory for RAG
Also in 4 projects that use this
JaimeGBASkillingest
Digest an external source (article, paper, transcript, URL, book chapter) and integrate its content into the knowledge base
Also in 8 projects that use this
ArtemXTechSkillquery
Query data from this vault. USE WHEN user asks about projects, clients, tasks, daily notes. Use grep to extract frontmatter - do NOT read full files.
Also in 4 projects that use this
szweibelSkillidea
Generate a wide field of options when the user is brainstorming, exploring, expanding, or pressure-testing — not recommending or implementing
Also in 3 projects that use this
GoogleSkilldashboard
Configures and launches the GKE Capacity & Triage Dashboard scoped to a GCP project ID, GKE cluster name, or target ML workload ID. Uses the configure-and-launch-dashboard action and launch-workload-dashboard workflow to verify local dashboard health, discover cluster capacity, trigger live scans, and generate deep-linked URLs. Use when the user types "/dashboard", says "launch dashboard", "open dashboard", "show capacity", or "view triage dashboard" with a project, cluster, or workload. Don't use for general workload log querying or automated failure attribution (use the troubleshoot skill).
Also in 4 projects that use this
Microsoft AzureSkilllint
Run all linting and spell checking for TypeScript and Rust code. Run this before committing to catch style issues, clippy warnings, and misspellings.
Also in 5 projects that use this
Other versions36 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 36
- AgriciDaniel/claude-obsidian★ 16ksame name
- egregore-labs/egregore★ 293same name
- romarayt/raytsystem-public-os★ 140same name
- thiagofinch/mega-brain★ 95same name
- sturlese/hippocampus★ 28same name
- endlessblink/master-plan★ 26same name
- pche-broken-artist/forge-of-thought★ 12same name
- CarbeneAI/Forge★ 9same name
- seongjaeryu/bindly-claude-code★ 8same name
- ARCASSystems/FounderOS★ 7same name
26 more not listed
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 repository24 other files it gives its agents.
- AGENTS.mdAGENTS.md
- agent-browserSkill
- applySkill
- browserSkill
- closeSkill
- connectSkill
- continueSkill
- dailySkill
- dream-memorySkill
- exploreSkill
- hand-overSkill
- improve-codeSkill
- lessonSkill
- memorySkill
- paste-this-into-your-aiSkill
- planSkill
- scheduleSkill
- servicesSkill
- shipSkill
- shopSkill
- update-dependenciesSkill
- verifySkill
- wong-setupSkill
- wong-syncSkill
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 1 is distinctive, and 1 is very specific to this project.
0
Common
Most of the 31 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 26
Check preconditions, then fetch main before comparing.
- Very specific: line 41
Keep the feature branch. On default/detached HEAD, git checkout -b "$SLUG": change/topic name, else worktree name for an unreadable session.
“Common” is measured: kept by at least 6 in 10 of the 31 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/save
/save authorizes branch creation, record upkeep, spec reconciliation, commit, push, PR updates and CI fixes; ask for anything else. Never force push, bypass hooks, amend merged commits or merge: /ship owns archive/merge, and the gate is never a local check.
/save [note]: a Status value sets Status; other notes seed the Decision log. Cold-resume context belongs in pushed files with repo-relative paths: another computer reads it.
1. Protect credentials and select the route
Keep every credential supplied, rotated, read or written this session out of tracked files, changes, facts, messages, PR text and reports. Values stay in working memory and their intended ignored live file; record only names and sources.
Load each matching procedure before its actions; conditions combine:
| Condition | Procedure |
| User supplied or rotated an explicitly named secret | Named-secret persistence, before writing records |
| Only facts, including a to-do that changed no repo file | Facts-only save |
| Code or a code plan needs a new change | New-plan fallback |
| The exact selected handoff is archived | The archived handoff |
Check preconditions, then fetch main before comparing; errors stop.
git fetch origin main
openspec context --json
openspec list --json
bash "$(git rev-parse --show-toplevel)/.claude/skills/save/scripts/change-candidates.sh" --jsonSave the change this work touched, found by the rungs, never a plan merely because it is the only one open. Keep BRANCH, NAME, CHANGE_ROOT separate; never rename or duplicate a change to match its branch.
A save changing repo files takes the normal route, except the validated record-only procedure above. Nothing learned, decided, or changed → report and stop.
2. Maintain the handoff and capture context
Keep the feature branch. On default/detached HEAD, git checkout -b "$SLUG": change/topic name, else worktree name for an unreadable session; append a short SHA on collision. Create required plan artifacts before checkpointing.
Maintain the living handoff:
- Proposal header: current
**Status:**, actual**Branch:**, and**Open questions:**(nonewhen empty). - Plan sections: the current intent from the latest agreed plan and diff, self-contained, edited in place under existing headings. Tasks: true checked state and agreed additions.
## Decision log: append one dated entry; never change earlier ones: a cold reader inherits the why from them.- Session facts: write them through the memory write gate with
"session": "current"and"source": "save". A spooled result does not block the save.
For an active change, reconcile its deltas. Read artifact and task progress from the CLI.
The archived handoff
Keep exact CHANGE_ROOT; infer NAME from its dated folder only if absent. Retain actual Branch and append a dated archive-checkpoint entry. The archive is the record: never rebuild the plan or recreate an active folder; archive already reconciled deltas. Stage archive, tracked removals and implementation; pass --mode archive and report its path.
3. Stage, exclude values, and commit
Stage intended implementation, handoff and removals, never git add .; check for unrelated staged work. A plan-only save commits its artifacts. Write, outside the repo, a one-line repo-style commit message ending Co-Authored-By: Claude, and the PR summary.
4. Publish and wait for the gate
One command runs the gate, review page to preview:
node "$(git rev-parse --show-toplevel)/.claude/skills/save/scripts/checkpoint.mjs" \
--message-file "$MESSAGE" --summary-file "$SUMMARY" \
--change-root "$CHANGE_ROOT" --mode active --scan-keys "$KEYS"--scan-keys names every credential handled this session; a match stops the save. No change → omit --change-root and --mode. Use the host's longest wait and wait again, silently, until it ends; then follow its NEXT: line.
5. Report
Report outcome and one link in plain words. Inside a verb or on request, include branch/commit, PR, change/archive Status, fact counts and stored/spooled/skipped state, CI fixes/uncertainty and preview/absence. Return the command's SAVE_HEAD, RECEIPT and SAVE_GATE_RESULT lines; always print the last inside a verb. Explicit errors; continue command only for active changes.
Direct invocation asks next steps with the question tool: continue, publish or stop; failed/unverified gates offer repairs.
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.

