deliver-change
Skill3 starsChanged 21 days agoMITView on GitHub
Implement a software change end to end, from plan to authorized draft pull request. Use for full implementation requests or implementation mode. Standalone phase requests use their own skill.
2 other projects use a version of this.
Adopted by
Most recently taken up by
- augchan42/inkstonetheir own version, first seen Nov 1, 2026
- hidekitux/skillstheir own version, first seen Sep 8, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- augchan42/inkstone★ 5same name
- hidekitux/skills★ 0same name
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 repository8 other files it gives its agents.
- implement-changeSkill
- plan-changeSkill
- publish-changeSkill
- remediate-findingsSkill
- review-changeSkill
- review-prSkill
- setup-projectSkill
- verify-changeSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
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 8
Run the full workflow within the user's scope.
- Distinctive: line 22
Run the phases in order. Continue only when the gate holds for the current candidate:
- Distinctive: line 42
Implement and verify: Run phases 2–3 per coherent slice, including required test-ID validation.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Deliver change
Run the full workflow within the user's scope. First read the contract core; each phase skill reads its own phase references.
1. Scope
- Identify the request, project authorities, workspace, explicit limits, and existing evidence. Reuse completed phases whose inputs remain valid.
- Standalone phase request → use that phase skill and stop at its boundary.
- Explicit limits such as no-commit or no-push stay binding in full delivery.
- Required skill or host capability unavailable → report the gap; never simulate it.
2. Phase gates
Run the phases in order. Continue only when the gate holds for the current candidate:
| # | Phase skill | Gate to continue |
| 1 | plan-change | Plan ready, material decisions resolved, implementation authorized |
| 2 | implement-change | Slice implemented, checks recorded |
| 3 | verify-change | Required checks pass |
| 4 | review-change | Fresh independent review is complete |
| 5 | remediate-findings, when findings need it | Corrections verified and re-reviewed |
| 6 | publish-change | Authorized operations done and read back |
A required check, review, or freshness result is missing, failed, unavailable, or inconclusive → stop the dependent phases, continue unaffected authorized work, and report the gap.
3. Phase inputs
- Plan. A revalidated accepted plan satisfies phase 1. Readiness does not approve an unresolved product or architecture choice.
- Implement and verify. Run phases 2–3 per coherent slice, including required test-ID validation. Missing validation mechanism → propose its scope, then add a small project-native check within implementation authority, or report a required gap.
- Review. Supply neutral context, current requirements, and the candidate, without a preferred verdict. Self-review never fills independent coverage.
- Remediate. Supply findings, candidate, requirements, edit authority, and the current round count and state. The three-round limit and earlier stopping conditions hold across calls and resumption. Accepting a real unresolved defect or a material redesign needs a decision.
- Publish. Supply the final candidate, current check and review evidence, dispositions, destination, operation permissions, explicit limits, prior attempts, and the criterion results updated with review evidence. Excluded operations stop; completed work stays. Mark ready only within authority, after required checks and review pass at the current revision.
4. Report
Report the behavior delivered, verification, findings, remaining gaps, and actual
publication state. Carry forward each criterion's status and gaps, and list
criteria whose required check did not run as Not verified. When blocked or
scope-limited, state what is complete and the next required action. Merge,
deployment, formal approval, and monitoring need separate authorization. These are
guided instructions, not controller-enforced gates.
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.