creating-change-records
Skill33 starsChanged 16 days agoMITView on GitHub
Creates or updates Quick or Standard change records, adds the required files, and refreshes what each claim must prove for an evidence-backed pull request. Use when starting or revising a change record. Do not use for a one-off throwaway script, or for work that belongs in an existing record rather than a new one.
In this repository34 other files it gives its agents.
- AGENTS.mdAGENTS.md
- challenge/AGENTS.mdAGENTS.md
- challenge/CLAUDE.mdCLAUDE.md
- CLAUDE.mdCLAUDE.md
- breaking-down-the-workSkill
- briefing-an-agentSkill
- checking-factsSkill
- checking-legal-and-safety-wordingSkill
- checking-release-readinessSkill
- checking-source-claimsSkill
- checking-what-a-change-affectsSkill
- choosing-what-to-controlSkill
- closing-stale-packetsSkill
- deciding-who-decidesSkill
- declaring-intentSkill
- double-checking-before-actingSkill
- handing-off-workSkill
- learning-from-experienceSkill
- organizing-project-foldersSkill
- proving-claimsSkill
- questioning-attitudeSkill
- rating-change-riskSkill
- recording-a-known-good-versionSkill
- recording-what-an-agent-didSkill
- reporting-shared-defectsSkill
- responding-to-incidentsSkill
- reviewing-code-qualitySkill
- running-a-review-panelSkill
- staying-on-missionSkill
- stress-testing-agent-changesSkill
- tracking-deficienciesSkill
- using-nuclear-gradeSkill
- verifying-final-artifactsSkill
- vetting-outside-code-and-modelsSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 3 are distinctive, and 1 is 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.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 18
Next action: add the missing file or link and rerun the validator.
- Distinctive: line 46
Add habit records (self-check, handoff, and the like) only when they change a decision or an action.
- Very specific: line 48
Keep status labels plain: pass, fail, gap, deferred, not applicable, or planned.
- Distinctive: line 69
Stop if the record turns into a long story with no links from claim to evidence.
≈ “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
Creating Change Records
Overview
A change record keeps the whole story in Git, together: the scope, what the change must do, the plan, the trace from claim to evidence, the proof, and the release decision. Use the smallest record that still lets a doubting reviewer decide.
Decision contract
- Claim checked: the record carries every file its mode requires, with the required links, exit criteria, source-lineage notes, and plain status labels (
pass,fail,gap,deferred,not applicable,planned) present and the internal links resolving. - Artifact observed: the mode from
risk.mdand thetemplates/quick/ortemplates/standard/templates -> a Quick record (risk.md,proof.md) or Standard record (risk.md,basis.md,plan.md,trace.md,verification.md,ship.md) plus the validator result. - Decision affected: block -- whether the packet is structurally complete and valid (
python tools/ng.py validatepass/fail). - Failure class: incomplete-record (a missing file, broken internal link, or proof command absent).
- Next action: add the missing file or link and rerun the validator; raise Quick to Standard or human review when proof cannot answer the reviewer.
When to Use
- You are starting a real AI-assisted change.
- You are updating a record after the scope, proof, risk, or release posture changed.
- You are preparing a pull request that needs evidence beyond the usual review notes.
- A preview, a self-check, a handoff, a lesson from operation (OPEX), or a supplier trust check is in play and needs a record.
When Not to Use
- The work leaves no lasting artifact and no need for review.
- The request is only to browse or explain existing docs.
- You only need to pick the mode / grade the change's risk, not author the record yet -- use
rating-change-riskinstead.
Inputs
- The chosen mode from
risk.mdor from the risk-rating skill. - A questioning-attitude screen, when uncertainty, AI power, dependency trust, or release stakes are real.
- Templates under
templates/quick/ortemplates/standard/. - The files the change affects and the proof commands.
- An existing record, if there is one.
Process
- Run
python tools/ng.py new <slug> --mode quick|standard. - Say, in your reply, which required files the chosen mode needs, by name -- not only in the files themselves, so a reader without tool access still sees the list. Name only the set for the mode in play: Quick needs
risk.mdandproof.md; Standard needsrisk.md,basis.md,plan.md,trace.md,verification.md, andship.md. - Fill in only the parts of each file that help someone decide.
- Add habit records (self-check, handoff, and the like) only when they change a decision or an action.
- Link to the files affected, the tests, the reviews, the source-map rows, and the known gaps.
- Keep status labels plain:
pass,fail,gap,deferred,not applicable, orplanned. - Run
python tools/ng.py validate .nuclear/changes/<slug>.
Outputs
- Quick record:
risk.md,proof.md. - Standard record:
risk.md,basis.md,plan.md,trace.md,verification.md,ship.md. - Habit records such as
turnover.md,self-check.md,opex.md, orsupplier-trust.mdwhen the stakes call for them. - The checker result.
Verification
- The required files exist.
- The required links, exit criteria, and source-lineage notes are present.
- The proof or verification file states the evidence status.
- The links inside the record resolve.
Escalation
- Move from Quick to Standard when Quick proof cannot answer the reviewer's question.
- Move to human review when a stronger written mode is in play.
- Stop if the record turns into a long story with no links from claim to evidence.
Common Rationalizations
- "We will fill it in after the pull request." The evidence should shape the work, not dress it up afterward.
- "Everything is obvious." If it matters to a future review, save the assumption and the path to the evidence.
- "One big record is easier." One change per record keeps the review small enough to do well.
Red Flags
- A missing proof command or evidence link.
- Claims that reach past what the tests or the review evidence cover.
- Record files that repeat each other instead of linking.
Prompt
Create or update a Nuclear-grade change record.
Inputs:
- slug: <slug>
- mode: <quick|standard>
- scope: <summary>
- affected files/assets: <list>
- questioned assumptions: <list>
- what the change must prove: <command/review/evidence>
- safety-habit (HPI) records started: <turnover/self-check/opex/supplier-trust/none>
Use the repo templates. Keep the record short. Lean on links, and point at evidence. Include the required links, the conditions for being done, and a note on where the ideas come from. Do not imply formal assurance or compliance.
Return:
- the required files for the chosen mode, named -- Quick: risk.md, proof.md; OR Standard: risk.md, basis.md, plan.md, trace.md, verification.md, ship.md (name only the set for the mode in play)
- the filled-in content for those filesSource-lineage note
This change-record skill is an original Git-native workflow. It draws on public configuration, lifecycle, assurance, secure-development, and release-readiness sources mapped in docs/00-standards-foundation/source-map.md. It does not create a certified quality assurance program.
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.

