pm-done
Skill2 starsChanged 8 months agoMITView on GitHub
Mark tasks, issues, or feature requests as completed and report cascading state changes. Automatically completes linked entities when a work package auto-completes.
In this repository15 other files it gives its agents.
- pm-auditSkill
- pm-brainstormSkill
- pm-cleanupSkill
- pm-codesmellsSkill
- pm-exploreSkill
- pm-housekeepingSkill
- pm-implementSkill
- pm-nextSkill
- pm-planSkill
- pm-refine-frSkill
- pm-scaffoldSkill
- pm-statusSkill
- pm-triageSkill
- pm-troubleshootSkill
- pm-verifySkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
15 rules. About 0 are common, 3 are distinctive, and 12 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.
12
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 44
Note: Do NOT prompt about missing rootCause/resolution.
- Very specific: line 64
Call mcp__pinkrooster__get_work_package_details with the WP ID
- Distinctive: line 66
If no non-terminal tasks found, report: "All tasks in {wpId} are already terminal."
- Very specific: line 71
If "Complete all" selected, call mcp__pinkrooster__batch_update_task_states with all task IDs
- Very specific: line 79
Check stateChanges for any entry where entityType is WorkPackage and newState is Completed
- Very specific: line 83
If the issue is NOT in a terminal state (Completed/Cancelled/Replaced):
- Very specific: line 84
Call mcp__pinkrooster__create_or_update_issue with projectId, issueId, and state: "Completed"
- Distinctive: line 87
If the FR is NOT in a terminal state (Completed/Rejected):
- Distinctive: line 118
If no cascading changes occurred, show: "No cascading state changes."
- Very specific: line 124
If completing a phase or WP and it has unverified acceptance criteria: "Verify criteria first?
- Very specific: line 125
If an auto-unblock occurred: "{entityId} was unblocked.
- Very specific: line 126
If a WP was auto-completed: "Work package complete.
- Very specific: line 127
If the completed WP had cancelled or replaced tasks: "WP had cancelled/replaced items.
- Very specific: line 128
If a single task was completed and the WP has remaining non-terminal tasks: "Continue implementing: /pm-implement {wpId}"
- Very specific: line 129
Always: after any completion: "Pick up next work: /pm-next" or "Check progress: /pm-status"
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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
Complete Work Items
Mark entities as completed and report all cascading state changes. Automatically propagate completion to linked entities without asking.
Determine Mode
Parse $ARGUMENTS to detect the mode:
- "all
<wp-id>" — Batch-complete all non-terminal tasks in a work package - Single entity ID — Complete one task, issue, or feature request
- Multiple entity IDs — Batch-complete multiple tasks (must be in same WP)
Detect entity type from ID format:
proj-N-wp-N-task-N→ Taskproj-N-issue-N→ Issueproj-N-fr-N→ Feature Requestproj-N-wp-N→ Work Package (used with "all" mode)
Mode: Single Task
- Call
mcp__pinkrooster__create_or_update_taskwith:taskId: the provided task IDstate:Completed
- Report the result and any
stateChangesfrom the response - Run Auto-Complete Linked Entities (see below)
Mode: Single Issue
- Call
mcp__pinkrooster__create_or_update_issuewith:projectId: extracted from the issue IDissueId: the provided issue IDstate:Completed
- Report the result
Note: Do NOT prompt about missing rootCause/resolution. The user can add those separately before calling /pm-done. Only ask if the user explicitly requests help documenting the fix.
Mode: Single Feature Request
- Call
mcp__pinkrooster__create_or_update_feature_requestwith:projectId: extracted from the FR IDfeatureRequestId: the provided FR IDstatus:Completed
Mode: Multiple Tasks (same WP)
- Extract the work package ID from the first task ID (strip
-task-N) - Call
mcp__pinkrooster__batch_update_task_stateswith:workPackageId: extracted WP IDtasks: array of{ taskId, state: "Completed" }for each provided task ID
- Report consolidated
stateChanges - Run Auto-Complete Linked Entities (see below)
Mode: "all <wp-id>"
- Call
mcp__pinkrooster__get_work_package_detailswith the WP ID - Collect all tasks that are NOT in a terminal state (Completed, Cancelled, Replaced)
- If no non-terminal tasks found, report: "All tasks in {wpId} are already terminal."
- Show the user which tasks will be completed using the
AskUserQuestiontool:- Question: "Complete these {N} tasks in {wpId}? {taskId1} "{name1}" ({state1}), {taskId2} "{name2}" ({state2}), ..."
- Header: "Batch done"
- Options:
[{label: "Complete all", description: "Mark all {N} non-terminal tasks as Completed"}, {label: "Cancel", description: "Do not complete any tasks"}]
- If "Complete all" selected, call
mcp__pinkrooster__batch_update_task_stateswith all task IDs - Report consolidated
stateChanges - Run Auto-Complete Linked Entities (see below)
Auto-Complete Linked Entities
This runs automatically after any task or batch completion that produces a WP auto-complete cascade. No user confirmation needed — the WP completing means the work is done.
- Check
stateChangesfor any entry whereentityTypeisWorkPackageandnewStateisCompleted - For each auto-completed WP:
a. Call
mcp__pinkrooster__get_work_package_detailswith the WP ID to getlinkedIssueIdsandlinkedFeatureRequestIdsb. For each issue ID inlinkedIssueIds: Callmcp__pinkrooster__get_issue_detailswith the issue ID- If the issue is NOT in a terminal state (Completed/Cancelled/Replaced):
- Call
mcp__pinkrooster__create_or_update_issuewithprojectId,issueId, andstate: "Completed" - Report: "Auto-completed linked issue {issueId} '{name}'"
c. For each FR ID in
linkedFeatureRequestIds: Callmcp__pinkrooster__get_feature_request_detailswith the FR ID
- Call
- If the FR is NOT in a terminal state (Completed/Rejected):
- Call
mcp__pinkrooster__create_or_update_feature_requestwithprojectId,featureRequestId, andstatus: "Completed" - Report: "Auto-completed linked feature request {frId} '{name}'"
- Call
- If the issue is NOT in a terminal state (Completed/Cancelled/Replaced):
Report Cascading State Changes
This is the most important output. Format cascades clearly:
## Completed
- {entityId} "{name}" -> Completed
## Cascading State Changes
(If stateChanges array is present and non-empty)
- **Phase auto-complete**: {phaseId} "{name}" -> Completed
(all tasks in phase reached terminal state)
- **WP auto-complete**: {wpId} "{name}" -> Completed
(all phases reached terminal state)
- **Auto-unblock**: {entityId} "{name}" -> {newState}
(blocker {blockerId} completed, restored from Blocked)
## Linked Entity Updates
(If any linked issues/FRs were auto-completed)
- **Issue auto-complete**: {issueId} "{name}" -> Completed
(linked WP {wpId} completed)
- **FR auto-complete**: {frId} "{name}" -> Completed
(linked WP {wpId} completed)If no cascading changes occurred, show: "No cascading state changes."
Suggest Next Steps
Choose the most relevant suggestions based on what happened:
- If completing a phase or WP and it has unverified acceptance criteria: "Verify criteria first?
/pm-verify {phaseId or wpId}— then re-run/pm-done" - If an auto-unblock occurred: "{entityId} was unblocked. Start it:
/pm-implement {entityId}" or "Pick up next priority:/pm-next" - If a WP was auto-completed: "Work package complete. Pick up next priority:
/pm-next" or "Check progress:/pm-status" - If the completed WP had cancelled or replaced tasks: "WP had cancelled/replaced items. Run
/pm-housekeepingto remove them." - If a single task was completed and the WP has remaining non-terminal tasks: "Continue implementing:
/pm-implement {wpId}" - Always after any completion: "Pick up next work:
/pm-next" or "Check progress:/pm-status"
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.

