Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/AIwork4me/open-llm-wikinpx agentmods add skills/aiwork4me/open-llm-wiki/query-writebackWrote 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/aiwork4me/open-llm-wiki/query-writeback)<a href="https://agentmods.dev/skills/aiwork4me/open-llm-wiki/query-writeback"><img src="https://agentmods.dev/badge/skills/aiwork4me/open-llm-wiki/query-writeback/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/aiwork4me/open-llm-wiki/query-writeback"><img src="https://agentmods.dev/badge/skills/aiwork4me/open-llm-wiki/query-writeback.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 | $0.00082 | $0.01142 |
| Opus 5.5 | $0.00033 | $0.00457 |
| Sonnet 5.5 | $0.00016 | $0.00228 |
| Haiku 4.5 | $0.00008 | $0.00114 |
Grade A, and why
query-writeback 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 1mo 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.
Query Writeback
Use the wiki to answer synthesis questions, then preserve high-value synthesis only when file changes are authorized.
Runtime Tools
Use deterministic scripts when available:
wiki_search.pyto find relevant source and concept pageswiki_claims.pyandclaims/claims.jsonlto inspect normalized durable claimswiki_normalize_metrics.pywhen comparisons depend on units, baselines, or protocolswiki_semantic_qa.pywhen the answer depends on claim-level evidence qualitywiki_science_review.pywhen a high-stakes synthesis needs human or second-LLM reviewwiki_writeback.pyto produce a reviewable diff before changing fileswiki_lint.pyafter approved writeback
Example:
uv run python scripts/wiki_search.py "<vault>" "<query>" --limit 8
uv run python scripts/wiki_claims.py "<vault>"
uv run python scripts/wiki_normalize_metrics.py "<vault>" --in-place
uv run python scripts/wiki_writeback.py "<vault>" --target concepts/<page>.md --query "<query>" --body-file <draft.md>
Trigger Boundary
Use this skill for questions that require cross-source synthesis:
- relationships between concepts
- comparisons across papers
- timelines or evolution paths
- recurring questions that would benefit from a reusable concept note
Do not use it for simple factual lookup, casual chat, non-wiki topics, or questions where a single source page already gives the answer.
File-Safety Boundary
- Read from the resolved wiki vault only.
- Answering is read-only by default.
- Writeback requires explicit user approval unless the current instruction has already authorized automatic wiki growth.
- Before writing, show the target files and a short summary of the proposed additions.
- Never rewrite a whole concept page when a targeted insertion is enough.
- Never update
sources/or QA reports from this skill. Usewiki-ingestfor source changes. - Always append a
log.mdentry for approved writeback.
Workflow
1. Resolve Scope
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.
- 1mo ago First seen · 145 lines · 82 tokens per session scan A c1e1aa2bf981
query-writeback is a skill published in the GitHub repository AIwork4me/open-llm-wiki (2 stars, last pushed 5mo ago), licensed MIT. It adds 82 tokens to every session and 1,142 once invoked, about $0.0003 per session on Opus 5.5. 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
llmwiki
Build and use a persistent, citation-traceable knowledge wiki from documents, research papers, notes, or project documentation. Use for llmwiki projects, retrieving reusable evidence for an agent task, or compiling a corpus for repeated questions. A one-off file read or live web search does not need a wiki.
llm-wiki
Build and maintain a persistent, interlinked Obsidian-compatible markdown wiki using Karpathy's LLM Wiki pattern. Extension-backed with auto-generated metadata, guardrails, and 14 custom tools (+3 opt-in agent-trajectory tools).
link-memory
Use after important user-approved decisions, when durable context should be proposed or reviewed, and for explicit Link memory lifecycle work: remember, recall, review, update, archive, restore, forget, or explain local memories through the CLI without requiring MCP.
link-retrieve
Use before answering work that may depend on user memory, project history, source-backed notes, or prior decisions; retrieve compact Link context through the CLI without loading the whole wiki or requiring MCP.
link-health
Use at the start of Link work when readiness is unclear, after installs or upgrades, and before repairs; verify health, inspect interrupted writes, back up, and repair generated indexes without MCP.
link-ingest
Use when raw files are present, source pages look stale, or a user asks to ingest notes into Link; refresh source-backed wiki pages, propose memories, and validate updates through the CLI without MCP.