papi-verify
Skill11 starsChanged 3 months agoMITView on GitHub
Use when the user says "verify", "health check", "check papi", "verify cycle", or after completing all cycle tasks before release. Runs a structured health check on the current PAPI cycle state: plan validity, build state, review coverage, branch hygiene, and data consistency.
In this repository14 other files it gives its agents.
- llms.txtllms.txt
- check-mcpSkill
- deployment-completeness-auditSkill
- designSkill
- papi-advancedSkill
- papi-autoSkill
- papi-buildSkill
- papi-ideaSkill
- papi-integration-recipeSkill
- papi-planSkill
- papi-strategySkill
- pr-reviewerSkill
- security-auditSkill
- usage-reporterSkill
The file
We split this skill into the rules it gives an agent
10 rules. About 0 are common, 4 are distinctive, and 6 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 12
Run a comprehensive verification of the current PAPI cycle state.
- Very specific: line 18
Run orient to get the current cycle number, task counts, and board state.
- Very specific: line 25
Run build_list to get all cycle tasks.
- Very specific: line 49
Run review_list to check:
- Very specific: line 57
Run git branch | grep feat/task and cross-reference with board state:
- Distinctive: line 66
Verify dashboard data matches MCP state:
- Distinctive: line 99
Run all checks even if early ones fail
- Very specific: line 100
Use PASS/WARN/FAIL consistently: PASS = no issues, WARN = minor issues that don't block, FAIL = something is wrong
- Distinctive: line 101
Always show the full report, even if everything passes
- Distinctive: line 102
If a check can't run (e.g., no In Progress tasks), report as PASS with a note
≈ “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
PAPI Verify — Structured Cycle Health Check
Run a comprehensive verification of the current PAPI cycle state. This is the "smoke test for PAPI itself" — catches regressions and data quality issues before they reach external users.
Steps
1. Orient and capture cycle state
Run orient to get the current cycle number, task counts, and board state. Record:
- Current cycle number
- Tasks by status (Backlog, In Progress, In Review, Done)
- Any carry-forward items
2. Plan validity check
Run build_list to get all cycle tasks. For each In Cycle task, verify:
- Task has a title (not empty)
- Task has a priority assigned
- Task has a complexity/effort estimate
- Task has a BUILD HANDOFF (check if it appears in build_list output)
Report: PLAN: X/Y tasks have valid handoffs
3. Build state check
For each task status, verify consistency:
In Progress tasks:
- Check if a
feat/task-XXXbranch exists (git branch | grep feat/task-XXX) - Flag any In Progress task with no branch (orphaned status)
Done tasks (this cycle):
- Check if a build report exists (these appear in the build_list or board_view output)
- Flag any Done task without a build report
Report: BUILD: X branches found for Y In Progress tasks. Z Done tasks with reports.
4. Review coverage check
Run review_list to check:
- How many tasks are In Review
- Are there Done tasks that were never reviewed (skipped review_submit)
Report: REVIEW: X tasks pending review, Y tasks reviewed and accepted
5. Branch hygiene check
Run git branch | grep feat/task and cross-reference with board state:
- Branches for Done tasks that haven't been merged (orphaned branches)
- Branches for tasks not in this cycle (leftover from previous cycles)
- Current branch — are we on main or a feature branch?
Report: BRANCHES: X feature branches, Y orphaned, Z ready to merge
6. Data consistency check
Verify dashboard data matches MCP state:
- Run
board_viewand compare task counts with orient output - Check if cycle number in orient matches what board_view shows
- Flag any discrepancies
Report: DATA: Cycle N, board shows X tasks, orient shows Y tasks
Output Format
Present results as a structured report:
PAPI VERIFICATION — Cycle N
=============================
PLAN: [PASS/FAIL] X/Y tasks with valid handoffs
BUILD: [PASS/FAIL] X branches for Y in-progress tasks, Z reports for W done tasks
REVIEW: [PASS/FAIL] X pending review, Y accepted
BRANCHES: [PASS/WARN/FAIL] X feature branches, Y orphaned
DATA: [PASS/FAIL] Counts consistent across orient and board_view
Overall: [PASS / X warnings / Y failures]
Failures:
- [list specific failures with task IDs]
Warnings:
- [list specific warnings]Rules
- This is READ-ONLY — do not fix anything, just report
- Run all checks even if early ones fail
- Use PASS/WARN/FAIL consistently: PASS = no issues, WARN = minor issues that don't block, FAIL = something is wrong
- Always show the full report, even if everything passes
- If a check can't run (e.g., no In Progress tasks), report as PASS with a note
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.

