AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Kb Lint

skill-pfeff-claude-skills-kb-lint · by pfeff

Report-only health check over the LLM Knowledge Base notes (reading the wider vault for connection candidates). Surfaces inconsistent/contradictory data, orphan notes, missing data web search could impute, and candidate new connections. Findings are report-only — any auto-fix is confined to KB-owned notes and reversible (git); off-limits and non-KB notes are proposals only. Operator-invoked via /…

No reviews yet
0 installs
36 views
0.0% view→install

Install

$ agentstack add skill-pfeff-claude-skills-kb-lint

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No 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.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-pfeff-claude-skills-kb-lint)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
2mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Kb Lint? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

KB Lint

On-demand health check over the LLM Knowledge Base (SPEC v2 §2.6). Reports issues; the operator applies them. The only autonomous writes allowed are reversible fixes to KB-owned notes; findings on off-limits zones or non-KB notes are surfaced as proposals only (INV-1).

> Status: the deterministic guards are implemented and unit-tested in kb-core > (is_kb_owned — auto-fix confined to KB notes, AC-6.2; is_writable — off-limits floor; > find_orphans — AC-6.3 input), plus kb-lint's own is_git_backed (re-checked fresh every > run against the vault's live git state; gates whether auto-fix is reachable at all — see > scripts/kb_lint_git.py). The findings pass is agent-orchestrated per operations/lint.md. > Proposal-only findings (declined within-KB link suggestions, KBnon-KB cross-links) are > persisted to a durable Notes/kb-lint-proposals.md inbox instead of only printing to the > transcript; the idempotency-key/dedup logic is implemented and unit-tested in > kb-lint/scripts/kb_lint_proposals.py.

Billing posture

/kb-lint is operator-invoked and interactive → subscription billing pool. No scheduled pass; same rationale as /kb-compile (SPEC v2 §6, CLAUDE.md billing guard).

What it reports (SPEC v2 §2.6)

  • Inconsistent / contradictory data across KB notes.
  • Orphan notes (no backlinks).
  • Missing data that a web search could impute.
  • Candidate new connections / articles (must surface ≥1 actionable candidate on a

non-trivial vault — AC-6.3). Proposal-only candidates (within-KB suggestions kb-lint declines to auto-apply, and KBnon-KB cross-links it can't write) are appended to a durable Notes/kb-lint-proposals.md inbox, deduped against prior runs.

Invocation

/kb-lint                  # health-check the KB notes

Execution

  1. Load the operation: Read(${CLAUDE_PLUGIN_ROOT}/skills/kb-lint/operations/lint.md)
  2. Execute it.
  3. Emit the findings report; apply only reversible fixes to KB-owned notes, surface the rest as proposals.

Integration Points

  • kb-core${CLAUDE_PLUGIN_ROOT}/skills/kb-core/scripts/kb_core.py (is_kb_owned + is_writable for the write guard, find_orphans for orphan detection).
  • kb-lint's own scripts${CLAUDE_PLUGIN_ROOT}/skills/kb-lint/scripts/kb_lint_git.py (is_git_backed — re-checked fresh every run, gates auto-apply eligibility); ${CLAUDE_PLUGIN_ROOT}/skills/kb-lint/scripts/kb_lint_proposals.py (idempotency key, checklist-line rendering, and dedup for the proposals inbox).
  • QMD — the vault search CLI, for connection-candidate discovery.
  • obsidian-notes skill / host config — vault path resolution; reads and KB-owned fixes (including the proposals inbox note).
  • kb-compile — produces the KB notes this skill audits.

Tests

cd skills/kb-lint/scripts && python3 -m unittest test_kb_lint_git test_kb_lint_proposals

Stdlib unittest (zero-dependency, matches kb-core/scripts/test_kb_core.py).

See Also

  • SPEC.md v2 §2.6 (lint contract) — workspace.
  • skills/kb-compile.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.