Install
$ agentstack add skill-nvzver-claude-marketplace-check ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
About
> Trace. On load, print first: =============== [manager/skills/check/SKILL.md] [manager] ===============
Check
Check the roadmap for hygiene issues — stale rows, missing pitches, status/branch mismatches — and apply the fixes the user approves. Dispatches the project-manager agent for the hygiene scan, gates each proposed row diff one by one, and re-renders the agent's applied rows. Does not contain hygiene-scan or roadmap-write logic — those live in the agent ([../../agents/project-manager.md](../../agents/project-manager.md) Steps 6-7). The dispatch → gate → re-render loop is the shared contract at [../../knowledge/roadmap-orchestration.md](../../knowledge/roadmap-orchestration.md).
Goal
Keep the roadmap honest — flag rows whose observable state contradicts their status and apply the corrections the user approves — without the user manually auditing the roadmap or invoking the agent directly.
Input
- None required. The roadmap is the entry point; the agent reads ambient state (roadmap, pitches, branches, specs) internally.
Steps
- Dispatch the project-manager and gate the row diffs — per the shared contract. Dispatch the
project-manageragent with an explicit intent (intent: hygiene-check) and run its returned gates per [../../knowledge/roadmap-orchestration.md](../../knowledge/roadmap-orchestration.md). The agent scans for hygiene issues per its own Steps 6-7 ([../../agents/project-manager.md](../../agents/project-manager.md) — the four scan conditions live there) and returns each finding as a proposed row diff — previous row + proposed row, each quoted withfile:line. This skill presents each diff one by one viaAskUserQuestion(approve / reject), sends approvals back viaSendMessagecontinuation, and re-renders each row the agent applies and quotes. Observable result: each hygiene diff gated; approved rows applied by the agent and re-rendered inline; rejected diffs discarded.
Output
Each proposed hygiene row diff is gated one by one; approved rows are applied by the agent (quoting each with file:line) and re-rendered inline by this skill; rejected diffs are discarded. If the roadmap is clean, the agent's confirmation is surfaced. No handoff. The agent proposes; this skill runs the gates and re-renders the writes.
Example Output
[illustrative]
Dispatching project-manager (intent: hygiene-check)...
Agent payload: 1 hygiene finding returned as a row diff.
Gate — .lsa/roadmap.yaml:21 onboarding-checklist (status backlog → in_progress; active feature/* branch exists): approve / reject
> approve
Applied — .lsa/roadmap.yaml:21 onboarding-checklist | Onboarding checklist | Should | in_progress
Constraints
- Orchestrator only — per [
../../knowledge/roadmap-orchestration.md](../../knowledge/roadmap-orchestration.md). Dispatch with an explicit intent, gate each diff, re-render. No hygiene-scan or roadmap-write logic here. - No silent handoff. The hygiene gates live in THIS skill (the agent cannot ask); the agent applies a row only after this skill returns its approval via continuation. No
lsa:discoverhandoff. - Show changes inline. Re-render each row the agent applies and quotes — never "roadmap updated" without the row. Per [
../../../core/skills/output/SKILL.md](../../../core/skills/output/SKILL.md) Rule 7. - Outputs follow [
core/output](../../../core/skills/output/SKILL.md) — citation by link, never restated.
/manager:check — manual invocation.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: NVZver
- Source: NVZver/claude-marketplace
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.