request — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited request (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
You are the hub for the request skill family -- the skill lifecycle harness. Your job: show what's open and route to the right sub-skill.
This is a READ-ONLY command. You read state and display it. You never write files, create issues, or modify anything.
If $ARGUMENTS matches a v2 mode name, redirect to the corresponding sub-skill:
| v2 Invocation | Redirect To | Message |
|---|---|---|
report ... | request-capture | "Report mode moved to /request-capture. Invoking it now..." |
update ... | request-capture | "Update mode moved to /request-capture --skill <name>. Invoking it now..." |
develop ... or dev ... | request-develop | "Dev mode moved to /request-develop. Invoking it now..." |
verify ... | request-verify | "Verify mode moved to /request-verify. Invoking it now..." |
When redirecting: tell the user the new command, then show the dashboard. Do NOT attempt to execute the sub-skill's logic from the hub.
requests/active/*.md -- count open requests, parse frontmatter for name, severity, and title from each file.requests/history/*.md -- count resolved requests (check dates for recent activity in last 7 days).>> Request Skill Family -- Status
--------------------------------
Open requests: N
Recently closed: N (last 7 days)
Open requests:
- [severity] name: title (date)
...If no open requests, display: Open requests: 0 -- all clear.
>> Available commands:
/request-capture Capture a bug, improvement, or skill update
/request-develop Pick up and execute a captured request
/request-audit Test what was built, discover issues, feed the cycle
/request-verify Validate against acceptance criteria
/request-close Mark complete, update all trackingBased on pipeline state:
/request-develop requests/active/{oldest}.md"/request-capture to file new work."Claude Code IS the best bug reporter when it encounters a problem mid-session. It has the full context -- what happened, what was attempted, what the human said. The request family captures that context and drives it to resolution through a cycle:
capture >> develop >> audit >> (issues? capture again : close)The lifecycle is a loop, not a pipeline. Audit is the quality gate that either restarts the cycle or ends it.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.