implement
Skill1 starsChanged 10 days agoMITView on GitHub
Develops a GitHub issue or idea until its requirements and fidelity criteria are verified. Use when asked to implement, build, or fix a described change or GitHub issue, or when the user runs /implement.
191 other projects use a version of this.
Adopted by
Most recently taken up by
- revisit-studies/studytheir own version, first seen Nov 9, 2026
- dz114879/skillstheir own version, first seen Nov 8, 2026
- Dimon94/dveritytheir own version, first seen Nov 8, 2026
- Dimon94/skillstheir own version, first seen Nov 8, 2026
- The-Vibe-Company/quivrtheir own version, first seen Nov 8, 2026
- miklschmidt/archboardtheir own version, first seen Nov 8, 2026
Used with
Projects with it also use setup-ts-deep-modules, ask-matt 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.
liufeng1976Skillsetup-ts-deep-modules
Wire dependency-cruiser into a TypeScript repo so each package is a deep module — implementation hidden in subfolders, reachable only through its entry-point files. User-invoked.
Also in 43 projects that use this
GoogleSkillask-matt
Ask which skill or flow fits your situation. A router over the skills in this repo.
Also in 91 projects that use this
RoentekSkillloop-me
Grill me about specs for the workflows I want to build, within this workspace.
Also in 50 projects that use this
AWSSkillto-questionnaire
Turn a decision you can't fully answer into a questionnaire for someone else to fill in.
Also in 88 projects that use this
ryotaroh180105Skillclaude-handoff
Hand the current conversation off to a fresh background agent that picks up the work immediately.
Also in 42 projects that use this
GoogleSkillto-spec
Turn the current conversation into a spec and publish it to the project issue tracker — no interview, just synthesis of what you've already discussed.
Also in 168 projects that use this
- +3 more from google
Other versions191 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 191
- mattpocock/skills★ 309ksame name
- QuivrHQ/quivr★ 40ksame name
- The-Vibe-Company/quivr★ 40ksame name
- foambubble/foam★ 17ksame name
- formbricks/formbricks★ 13ksame name
- ArcReel/ArcReel★ 6.3ksame name
- vinvcn/mattpocock-skills-zh-CN★ 5.5ksame name
- fallow-rs/fallow★ 5.3ksame name
- withcoral/coral★ 4.9ksame name
- bestofjs/bestofjs★ 3.1ksame name
181 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 repository20 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- address-feedbackSkill
- codebase-auditSkill
- code-reviewSkill
- create-issueSkill
- create-prSkill
- deliverSkill
- delivery-waitSkill
- domain-modelingSkill
- git-workflowSkill
- grillingSkill
- grill-with-docsSkill
- improve-codebase-architectureSkill
- native-nostalgia-stackSkill
- project-structureSkill
- roadmap-reviewSkill
- software-engineering-excellenceSkill
- test-against-specSkill
- update-prSkill
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 4 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 171 edited versions we read keep these.
4
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 39
Read project instructions, relevant code, tests and durable decisions.
- Distinctive: line 43
Run the named reproduction or inspect the artifact.
- Distinctive: line 54
If required external evidence is unavailable, stop the dependent work.
- Distinctive: line 69
Run focused developer checks while fixing observed gaps.
“Common” is measured: kept by at least 6 in 10 of the 171 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
Implement
Develop the smallest complete change that satisfies the agreed requirements. Reuse settled scope, approach and authorization. A failed check, diagnosis or available fix does not end implementation. Only a material unresolved decision, new authority or an external blocker after safe alternatives are exhausted can stop dependent work; complete independent work before asking. If a skill requires a pause, link its loaded file as a Markdown link and quote the rule verbatim in a block quote; a paraphrase or a bare path does not let the user check the rule.
Establish the work
Infer the target from the request, conversation, handoff and repository; no issue/idea selector is required. Verify an identified issue's repository, requirements, comments, readiness and current state. A failed lookup does not make it an unfiled idea. For a described outcome, search for matching work and reuse its decisions. Otherwise establish a concise mini-spec with the outcome, scope, constraints and testable success measures; do not file an issue merely to implement it. Resolve conflicting or ambiguous targets before editing.
Read project instructions, relevant code, tests and durable decisions. Use the nearest Definitions of Ready and Done; if absent after a real search, use the workflow's checks and declared project commands and record the acceptance gap. Inspect the selected toolkit's overall capabilities before adding machinery. Run the named reproduction or inspect the artifact. If an issue is already fixed, verify its current behavior and regression coverage rather than duplicate it; add missing coverage when that is the remaining requirement.
Reuse a settled approach after checking it against current evidence. Research
primary sources when a decision depends on current external facts and reconcile
them with the installed version. For an unresolved material choice or requested
comparison, read references/approach-selection.md.
It owns comparison evidence and the registered grilling loop. Explicit
automatic mode skips that interview while preserving investigation and the
user's ownership of material product, architecture, security or scope choices.
If required external evidence is unavailable, stop the dependent work; continue
independent investigation and any established in-scope correction. A
conclusively failed readiness threshold blocks dependent work the same way.
Develop and verify
This loop covers development only: /deliver owns end-to-end delivery and
/create-pr owns publication. Reuse the confirmed requirements and approach.
- Reuse or create a focused branch/worktree under
git-workflow. Apply its clean-worktree, fresh-base and default-branch health gates before new work; preserve owned in-progress changes when resuming the same implementation. - Implement the smallest complete change, then run and inspect the real result. For UI/UX work, compare affected states with the requested appearance, interactions and fidelity; include relevant accessibility and viewports. Run focused developer checks while fixing observed gaps.
- Apply
/code-review fix-alland/test-against-spec fix. Review the actual change against the agreed requirements; test observable behavior against explicit specification sources. Use their direct documented equivalents if unavailable. Reuse matching evidence without repeating already-covered work. - Repair every verified requirement gap, including visual or behavioral fidelity, regardless of a finding's severity. Optional improvements outside the agreed outcome do not become requirements. Any edit invalidates affected evidence: repeat the necessary review, inspection and behavior checks until they agree on the unchanged result. Own the final applicable Definition of Done and project gate; run only its missing or invalidated checks.
Continue safe in-scope repair; reconsider an approach that is not advancing acceptance. If required behavior needs an unavailable exact-revision preview, return the specific publication need to the active delivery or PR caller, which owns that operation and resumes testing afterward. A standalone implementation asks only for the missing authority or environment.
Finish
Finish when every verified requirement gap is resolved and applicable gates pass; unrelated improvements do not extend the task. Return the implemented result and observed requirement evidence to the caller. Report unresolved required behavior as incomplete; never waive it because its finding is low severity.
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.

