38
GitHub Stars
2
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill khgs2411/flow --skill flow-verifier- SKILL.md19.4 KB
- VERIFY.md4.4 KB
Overview
This skill verifies plan consistency, generates concise project summaries, and helps maintain plan health with safe maintenance operations. It focuses on read-only inspection for reviews and audits, and includes controlled maintenance actions like compacting and rollback with backups. Use it to detect structure, status, and reference issues so teams can correct the source of truth quickly.
How this skill works
The verifier reads plan files (DASHBOARD.md, PLAN.md, phase and task files) and runs pattern-based checks for status marker consistency, phantom tasks, implementation gate violations, task structure, action-item completion, and file reference validity. It reports structured findings with severity levels and actionable recommendations. For maintenance it creates backups, compacts old completed work, and can roll back to recent backups when requested.
When to use it
- Before stakeholder reports or sprint demos to ensure accuracy
- When you want a full plan review or a task-specific audit
- To validate status markers across phases and tasks
- Before running a compact or rollback operation
- To generate a project summary for meetings
Best practices
- Run a full plan review after each sprint or major update
- Keep brainstorming complete before implementation to avoid gate violations
- Limit IN PROGRESS to one active item to avoid context confusion
- Auto-backup before any compact or destructive maintenance
- Treat phantom tasks as critical — remove or create missing files promptly
Example use cases
- Full plan review: scan DASHBOARD.md, phases, tasks, and iterations to produce a consolidated report
- Task-specific audit: verify a single task for structure, markers, action items, and file references
- Status audit: grep all status markers, flag lifecycle violations and multiple IN PROGRESS items
- Summary generation: produce a project summary (overview, phase progress, next steps) for stakeholder updates
- Compact operation: prune old completed iteration details while preserving outcomes and timestamps
FAQ
No. Read-only checks never modify files. Maintenance commands (compact, rollback) only run when explicitly requested and always create a backup first.
What counts as a critical issue?
Critical issues include phantom tasks, implementation starting before brainstorming completes, and task structure violations (tasks containing both direct action items and iterations). These should be fixed before proceeding.