Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add po4yka/llm-wiki-skills --skill llm-wiki-trust-auditgit clone --depth 1 https://github.com/po4yka/llm-wiki-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/po4yka/llm-wiki-skills/llm-wiki-trust-audit)<a href="https://agentmods.dev/skills/po4yka/llm-wiki-skills/llm-wiki-trust-audit"><img src="https://agentmods.dev/badge/skills/po4yka/llm-wiki-skills/llm-wiki-trust-audit/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/po4yka/llm-wiki-skills/llm-wiki-trust-audit"><img src="https://agentmods.dev/badge/skills/po4yka/llm-wiki-skills/llm-wiki-trust-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00081 | $0.00919 |
| Opus 5 | $0.00041 | $0.00460 |
| Sonnet 5 | $0.00016 | $0.00184 |
| Haiku 4.5 | $0.00008 | $0.00092 |
Grade A, and why
llm-wiki-trust-audit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.
LLM-Wiki Trust Audit
Goal
Evaluate whether an LLM-Wiki is trustworthy enough to use for decisions and identify the highest-risk failure modes.
When to use
- The user asks whether their LLM-Wiki can be trusted before relying on it for a decision.
- Before granting an agent write access to a vault, or after granting it, to confirm the write-safety boundary still holds.
- After a large ingest, bulk edit, or lint run, to confirm provenance and review gates were not weakened.
- Route setup/permission-only reviews to
llm-wiki-security-reviewand proposal-specific risk reviews tollm-wiki-critique-auditinstead of this skill.
Inputs
- Vault/repository path.
AGENTS.md,CLAUDE.md, skills and schemas.raw/,wiki/,_meta/,_agent/reports/if present.- Git history if available.
- User's risk tolerance and domain sensitivity.
Procedure
1. Inspect the trust model
Check whether the system defines:
- raw source immutability;
- page lifecycle states;
- source backlinks;
- claim types;
- confidence semantics;
- review requirements;
- protected human sections;
- lint cadence;
- rollback/recovery path.
2. Audit provenance
Sample important pages and report:
- no provenance;
- source-level provenance only;
- claim-level provenance;
- generated pages citing generated pages;
- stale or missing source hashes;
- source links that no longer resolve.
Re-verify any page flagged stale or with an unresolved source link before treating its claims as trustworthy.
3. Audit generated content boundaries
Look for:
- AI-generated pages marked reviewed/verified without evidence;
ai_confidencedefaulted lazily;- low-confidence pages not requiring review;
- ambiguous claims in trusted pages;
- human synthesis sections overwritten or unprotected.
4. Audit structural health
Coordinate with wiki-lint if available. Check:
- broken links;
- orphan pages;
- duplicate concepts;
- taxonomy drift;
- stale pages;
- contradiction reports ignored;
- excessive draft backlog.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 145 lines · 81 tokens per session scan A c5be5203a242
llm-wiki-trust-audit is a skill published in the GitHub repository po4yka/llm-wiki-skills (3 stars, last pushed 19d ago), licensed MIT. It adds 81 tokens to every session and 919 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
okf-harness-bootstrap
This unified entrypoint routes setup and workspace maintenance without making the user choose a mode.
okf-harness
One Door routes OKF Harness workspace requests to exactly one internal workflow at a time.
wiki-lint
Health-check a wiki vault. Finds orphan pages (no inbound links), dead wikilinks (point to non-existent pages), missing frontmatter fields, stale claims, empty sections, and pages absent from catalog.md. Produces a structured report with severity tiers and proposes concrete fixes — but does not auto-apply them unless…
meta-sync-template
Interactive flow to propagate the reference (.template) vault's plugins, snippets, and root docs to one or more configured vaults. Reads the active router config, lists every vault with online status, flags vaults missing obsidian-local-rest-api upfront, lets the user pick all or a subset, and optionally re-clones…
save
File the current Claude conversation (or a specific insight from it) as a structured wiki note. Auto-detects the right type (decision, answer, session-log, technique, ADR), writes appropriate frontmatter, places the file in the correct wiki folder, and updates catalog.md, journal.md, hot.md. Use when the user says…
wiki-boundary
Rank the "frontier" pages of a wiki — the crossroads many pages link to that stay thin inside — so you can decide where research would pay off most. Use when the user says "what should I write about next", "where are the gaps", "which pages need work", "find the thin hub pages", "boundary pages", "/wiki-boundary"…