change
Skill40 starsChanged 4 months agoMITView on GitHub
Handle mid-track requirement changes. Analyzes impact on completed and pending tasks, proposes amendments to spec.md and plan.md before applying.
9 other projects use a version of this.
Adopted by
Most recently taken up by
- dmitribogatenkov/briefprooftheir own version, first seen Nov 8, 2026
- sanjeevkumarraob/chartertheir own version, first seen Oct 25, 2026
- formal-you/agent-skills-suitetheir own version, first seen Oct 20, 2026
- iadr-dev/greenlighttheir own version, first seen Oct 20, 2026
- RTekaya/app-factorytheir own version, first seen Oct 15, 2026
- luiscortex/superpowers-freeze-specstheir own version, first seen Oct 11, 2026
Other versions9 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 9
- microsoft/fluentui★ 20ksame name
- dmitribogatenkov/briefproof★ 47same name
- samson-art/transcriptor-mcp★ 24same name
- iadr-dev/greenlight★ 2same name
- benbatya/claude-skills★ 0same name
- formal-you/agent-skills-suite★ 0same name
- luiscortex/superpowers-freeze-specs★ 0same name
- RTekaya/app-factory★ 0same name
- sanjeevkumarraob/charter★ 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 repository35 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- web/llms-full.txtllms.txt
- web/llms.txtllms.txt
- adrSkill
- assist-reviewSkill
- bughuntSkill
- coverageSkill
- debugSkill
- decomposeSkill
- deep-reviewSkill
- deploy-checklistSkill
- discoverSkill
- docsSkill
- documentationSkill
- draftSkill
- graphSkill
- impactSkill
- implementSkill
- incident-responseSkill
- initSkill
- integrationsSkill
- jiraSkill
- learnSkill
- new-trackSkill
- opsSkill
- planSkill
- quick-reviewSkill
- revertSkill
- reviewSkill
- standupSkill
- statusSkill
- tech-debtSkill
- testing-strategySkill
- tourSkill
- uploadSkill
The file
We split this skill into the rules it gives an agent
10 rules. 0 are common, 2 are distinctive, and 8 are very specific to this project.
0
Common
Most of the 8 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
8
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 27
If draft/ does not exist: STOP — "No Draft context found.
- Very specific: line 40
If no track <id> specified:
- Very specific: line 43
If no [~] track, find the first [ ] Pending track
- Distinctive: line 46
If no change description provided:
- Very specific: line 54
Read draft/tracks/<id>/spec.md — extract requirements and acceptance criteria
- Very specific: line 55
Read draft/tracks/<id>/plan.md — extract all tasks with their current status ([ ], [~], [x], [!])
- Very specific: line 56
Read draft/tracks/<id>/metadata.json — for track type and status
- Very specific: line 57
Read draft/tracks/<id>/hld.md if present — extract Architecture, Detailed Design components, Dependencies, Checklist sections, and …
- Very specific: line 58
Read draft/tracks/<id>/lld.md if present — extract Classes/Interfaces, Data Model, Algorithms, Error Handling sections
- Distinctive: line 103
Re-approval flag: If the HLD Approvals table has any signed rows (Date column populated) AND the change touches HLD structural sections …
“Common” is measured: kept by at least 6 in 10 of the 8 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
Course Correction
You are handling a mid-track requirement change using Draft's Context-Driven Development methodology.
Red Flags - STOP if you're
- Applying changes to spec.md or plan.md without showing the user what will change first
- Invalidating
[x]completed tasks without flagging them explicitly - Proceeding past the CHECKPOINT without user confirmation
- Editing files when the user said "no" or "edit"
Show impact before applying. Always confirm.
Step 0: Verify Draft Context
ls draft/tracks.md 2>/dev/nullIf draft/ does not exist: STOP — "No Draft context found. Run /draft:init first."
Step 1: Parse Arguments
Extract from $ARGUMENTS:
- Change description — free text describing what needs to change (required)
- Track specifier — optional
track <id>prefix to target a specific track
Default Behavior
If no track <id> specified:
- Auto-detect the active
[~]In Progress track fromdraft/tracks.md - If no
[~]track, find the first[ ]Pending track - Display:
Auto-detected track: <id> - <name>before proceeding
If no change description provided:
- Error: "Usage:
/draft:change <description>or/draft:change track <id> <description>"
Step 2: Load Context
- Read
draft/tracks/<id>/spec.md— extract requirements and acceptance criteria - Read
draft/tracks/<id>/plan.md— extract all tasks with their current status ([ ],[~],[x],[!]) - Read
draft/tracks/<id>/metadata.json— for track type and status - Read
draft/tracks/<id>/hld.mdif present — extract Architecture, Detailed Design components, Dependencies, Checklist sections, and Approvals table (signed/unsigned) - Read
draft/tracks/<id>/lld.mdif present — extract Classes/Interfaces, Data Model, Algorithms, Error Handling sections
Step 3: Analyze Spec / HLD / LLD Impact
Analyze the change description against the loaded spec, HLD, and LLD.
Code-grounded impact (Ground-Truth Discipline G1, G2, G4): Classification ("Modified", "Unaffected", etc.) must be informed by the current code, not just the prior spec text. For each requirement / AC you're about to mark Unaffected, confirm the code path it depends on still behaves as the spec claims — Read the cited file or a representative file in the affected module before stamping Unaffected. Specs and implementations drift; "spec text unchanged" ≠ "behavior unchanged."
For each requirement and acceptance criterion, classify the effect:
| Classification | Meaning |
| Added | New requirement or AC introduced by this change |
| Modified | Existing requirement or AC needs updating |
| Removed | Existing requirement or AC is no longer needed |
| Unaffected | No change needed |
Produce a concise impact list. Example:
Spec impact:
- AC #2 "User can export to CSV" → Modified (now also requires JSON format)
- AC #5 "Export limited to 1000 rows" → Removed (no row limit)
- NEW: AC #6 "Export progress indicator for large datasets"HLD impact (only when hld.md exists):
- §Architecture / Component Diagram — does the change introduce new modules or alter integration edges?
- §Detailed Design — does any per-component subsection need updating, or are new components introduced?
- §Dependencies — new/removed dependent components?
- §Checklist (Performance, Scale, Security, Resiliency, Multi-tenancy, Upgrade, Cost) — do any answers need re-evaluation?
- §IP / TPT — does the change introduce new third-party technology or invention disclosure?
- §Deployment — does the deployment surface change?
LLD impact (only when lld.md exists):
- §Classes and Interfaces — signatures added/modified/removed?
- §Data Model — schema changes? New fields? Migration required?
- §Key Algorithms and Workflows — algorithm changes? New sequence diagrams needed?
- §Error Handling — new error classes or retry policy changes?
- §Observability — new metrics or alert thresholds?
Re-approval flag: If the HLD Approvals table has any signed rows (Date column populated) AND the change touches HLD structural sections (Architecture, Detailed Design, Dependencies, Checklist, IP, Deployment), surface this warning prominently:
⚠️ HLD modified after sign-off — Approvals table requires re-circulation.
Signed rows: [list which roles signed and when]
Changed sections: [list of HLD sections impacted]Same logic applies to LLD Approvals when LLD §Classes/Interfaces, §Data Model, or §Key Algorithms change.
Step 4: Map Impact to Plan Tasks
For each task in plan.md, determine if the spec change affects it:
[x]completed tasks that are now invalidated by the change → flag as:⚠️ [task description] — may need rework[ ]pending tasks that need updating → show the proposed new task text[~]in-progress tasks that are affected → flag as:⚠️ IN PROGRESS: [task description] — review before continuing[!]blocked tasks that are affected → flag as:⚠️ BLOCKED: [task description] — re-evaluate; requirement change may alter blocking condition or resolution path- Unaffected tasks — skip, do not mention
Step 5: Present Impact Summary
Display a clear summary before proposing any file changes:
Change: [change description]
Track: <track_id> — <track_name>
Spec impact:
- [classification] [requirement/AC]
- [classification] [requirement/AC]
Plan impact:
- ⚠️ [N] completed task(s) may need rework
- [M] pending task(s) need updating
- [K] in-progress task(s) need review
- [B] blocked task(s) need re-evaluation
Completed tasks that may need rework:
- [x] [task description] (commit: abc1234)
Pending tasks with proposed changes:
Before: - [ ] [original task text]
After: - [ ] [proposed new task text]Step 6: Show Proposed Amendments
Display only the changed sections of each file (not full rewrites):
Proposed spec.md changes
Show the diff as before/after for each modified section. Do not rewrite unchanged sections.
Proposed plan.md changes
Show each task that would be modified as before/after. Do not rewrite the full plan.
Proposed hld.md changes (when HLD exists and is impacted)
Show before/after for each impacted HLD section. Preserve §Approvals table verbatim — do not modify Approvals as part of the spec amendment; re-circulation is the author's manual step. Surface a "Sections changed" list at the top of the diff so reviewers can see scope at a glance.
Proposed lld.md changes (when LLD exists and is impacted)
Show before/after for each impacted LLD section. Preserve §Approvals verbatim. Flag schema changes (Data Model) for migration consideration.
Step 7: CHECKPOINT
Apply these changes to spec.md and plan.md? [yes / no / edit]yes— proceed to Step 8no— discard all proposed changes, announce "No changes applied." and stopedit— let the user describe adjustments to the proposed amendments, then revise and re-present the CHECKPOINT again. The loop continues until the user selectsyesorno.
Step 8: Apply Changes and Log
- Apply the agreed amendments to
spec.md,plan.md, and (when impacted)hld.md/lld.md. Preserve §Approvals tables in HLD/LLD verbatim — re-approval is a manual author step, not an automated edit. - Update
draft/tracks/<id>/metadata.json:- Set
updatedto current ISO timestamp - Recalculate
tasks.totalby counting all- [ ],- [~],- [x], and- [!]lines in the updatedplan.md. Updatetasks.completedby counting only- [x]lines.
- Set
- Append a Change Log entry (with current git SHA (obtain via
git rev-parse --short HEAD) and timestamp) toplan.md. If a## Change Logsection does not exist, add it at the bottom:
## Change Log
| Date | Description | Impact |
|------|-------------|--------|
| [ISO date] | [change description] | [N completed may need rework, M pending updated] |- Announce:
Changes applied: <track_id>
Updated:
- draft/tracks/<id>/spec.md
- draft/tracks/<id>/plan.md
[when HLD impacted:]
- draft/tracks/<id>/hld.md
[when LLD impacted:]
- draft/tracks/<id>/lld.md
[If completed tasks flagged:]
⚠️ Review N completed task(s) — they may not align with the updated spec.
Re-run /draft:implement to address rework, or /draft:review to assess.
[If HLD/LLD modified after sign-off:]
⚠️ HLD/LLD modified after sign-off — re-circulate to approvers listed in §Approvals.
/draft:upload will block git upload for high-criticality tracks until re-signed.
Next: /draft:implement to continue, or /draft:review to assess current state.Error Handling
Track Not Found
Error: Track '<id>' not found.
Run /draft:status to see available tracks.No Active Track
Error: No active track found.
Use: /draft:change track <id> <description>No Spec or Plan
Error: Missing spec.md or plan.md for track <id>.
Cannot perform change analysis without both files.Examples
Change description for active track
/draft:change the export format should support JSON in addition to CSVTargeting a specific track
/draft:change track add-export-feature also require a progress indicator for exports over 500 rowsDoes 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.

