spike
Skill11 starsChanged 2 days agoApache-2.0View on GitHub
Opens a throwaway spike of this project in its own git worktree, beside the project directory, with every hook off there except the audit trail — no format, check, tests, compose gate or guard — so exploratory code can be written fast while the main checkout stays fully enforced and the spike's token spend still shows in /audit-usage. Explicit invocation only.
56 other projects use a version of this.
Adopted by
Most recently taken up by
- ChristopherAlphonse/calphonse-skillstheir own version, first seen Nov 16, 2026
- rodos4312/Meu-Agenttheir own version, first seen Nov 15, 2026
- fabioc-aloha/Alex_ACT_ONEtheir own version, first seen Nov 15, 2026
- metaspartan/cybaratheir own version, first seen Nov 15, 2026
- NousResearch/hermes-agenttheir own version, first seen Nov 13, 2026
- hnqlv/skillstheir own version, first seen Nov 12, 2026
Used with
Projects with it also use competitor-news-monitor, blocked-page-recovery and 10 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.
Samin12Skillcompetitor-news-monitor
Watch named companies for material news; cited digests.
Also in 6 projects that use this
nobodyohm-webSkillblocked-page-recovery
Recover blocked/paywalled/WAF'd pages via fallbacks.
Also in 6 projects that use this
Samin12Skilldocument-to-action-items
Extract cited obligations, deadlines, tasks from documents.
Also in 6 projects that use this
michaellam26Skillsdlc-review
Trigger stage-specific reviews for an SDLC project and save results to reviews/
Also in 5 projects that use this
john-data-chenSkillteams-meeting-pipeline
Operate the Teams meeting summary pipeline via Hermes CLI — summarize meetings, inspect pipeline status, replay jobs, manage Microsoft Graph subscriptions.
Also in 8 projects that use this
Tommy-ywSkillsongwriting-and-ai-music
Songwriting craft, AI music generation prompts (Suno focus), parody/adaptation techniques, phonetic tricks, and lessons learned. These are tools and ideas, not rules. Break any of them when the art calls for it.
Also in 8 projects that use this
- +1 more from Samin12
Other versions56 projects use a version of this.
Word for word · 1
- ice-lfernandes/claude-spring-architect★ 11100% same
Edited · 0
None found.
Wrote their own version · 55
- molt-bot/clawdbot★ 395ksame name
- steipete/warelay★ 395ksame name
- moltbot/clawdbot★ 395ksame name
- openclaw/openclaw★ 395ksame name
- steipete/clawdis★ 394ksame name
- moltbot/moltbot★ 394ksame name
- clawdbot/clawdbot★ 394ksame name
- NousResearch/hermes-agent★ 264ksame name
- OpenBMB/PilotDeck★ 5ksame name
- friuns2/openclaw-android-assistant★ 573same name
45 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 repository35 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- ai-architectSkill
- arch-adoptSkill
- arch-doctorSkill
- audit-usageSkill
- cache-architectSkill
- claude-code-architect-designerSkill
- docker-architectSkill
- domain-modelingSkill
- git-publishSkill
- gof-design-patternsSkill
- graphql-api-architectSkill
- harness-lessonsSkill
- http-client-architectSkill
- init-projectSkill
- jobs-architectSkill
- messaging-architectSkill
- nerviz-diagramsSkill
- new-feature-implementSkill
- new-featureSkill
- new-projectSkill
- persistence-architectSkill
- platform-upgradeSkill
- project-bootstrapSkill
- project-configSkill
- report-issueSkill
- rest-api-architectSkill
- search-architectSkill
- security-architectSkill
- sonar-lessonsSkill
- sonarqube-setupSkill
- test-architectSkill
- transport-security-setupSkill
- triage-issueSkill
- use-case-designSkill
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 0 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 51 edited versions we read keep these.
0
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 36
Preconditions: Run from the main checkout: git rev-parse --git-dir equals git rev-parse --git-common-dir.
- Very specific: line 50
Not the main checkout: ❌ /spike runs from the main checkout.
“Common” is measured: kept by at least 6 in 10 of the 51 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
spike — a throwaway worktree where only the audit trail stays on
/spike <slug> creates ../<project>-spike-<slug> on branch spike/<slug> and reduces that
worktree's .claude/settings.json to the hooks spike.keep_modes lists. The main checkout
never relaxes, and this session stays enforced; the person opens claude in the worktree.
Why this is a skill (Form 2) and not an agent
Form 2 because the trigger is the person deciding to explore (axis 2) and its effect — a worktree with enforcement off — must never start on the model's own decision. It is not an agent: none of the three reasons applies, the output is two lines and the work is three commands. Not disableAllHooks in a settings.local.json, which turns the audit trail off with everything else (0198, option 2).
Contract
- Class: ops —
skill_classes.opsin@.claude/schemas/extensions.json: writes no file of this project with a tool; its effect is a sibling worktree, made bygit worktree addandArchHook.jar spike. - Reads the project root (
git rev-parse --show-toplevel) and the argument. - Writes nothing inside the project. Outside it: the worktree
../<project>-spike-<slug>on branchspike/<slug>, whose.claude/settings.jsonArchHook.jar spikereduces to thespike.keep_modesentries and marksskip-worktree. - Does not merge, commit or push the spike, nor delete it —
endprints the commands. Promoting spike code is a/new-featurerun in the main checkout. - Design:
@.claude/decisions/0198-spike-worktree-with-hooks-off.md.
Procedure
- Preconditions. Run from the main checkout:
git rev-parse --git-direqualsgit rev-parse --git-common-dir; otherwise stop (failure mode 1). The argument is a slug matching[a-z0-9][a-z0-9-]*; missing → ask once withAskUserQuestion, then stop. end <slug>— print, do not run:git worktree remove ../<project>-spike-<slug>andgit branch -D spike/<slug>, plus one line: the spike's spend is already in/audit-usage. Stop.- Create.
git worktree add ../<project>-spike-<slug> -b spike/<slug>, where<project>is the basename of the root. Exists already → failure mode 2. - Relax.
java -jar .claude/hooks/ArchHook.jar spike ../<project>-spike-<slug>and relay its stdout line verbatim. Non-zero → failure mode 3. - Report, then stop:
✅ Spike <slug> — ../<project>-spike-<slug> on spike/<slug> Hooks ....... audit only (format, check, tests, compose gate, guard off there) Next ........ open `claude` in that directory; this session stays enforced End ......... /spike end <slug>
Failure modes
- Not the main checkout:
❌ /spike runs from the main checkout; this is a linked worktree. Open the project root and run it there. - Worktree or branch exists:
❌ ../<project>-spike-<slug> or spike/<slug> already exists. Pick another slug, or /spike end <slug> first. spikemode refused:❌ ArchHook spike refused: <its stderr first line>. The worktree exists with every hook on — remove it with git worktree remove ../<project>-spike-<slug>.
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.

