query

A command for answering questions using information stored in a project wiki.

In plain words
What is it for?
It is for researching a question in the local wiki and, when requested, saving the answer as a dated synthesis page.
Why use it?
It gathers relevant wiki pages and cites them, making answers easier to verify and highlighting missing or conflicting information.

Command

Install

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.

agentmods
npx agentmods add commands/doneyli/claude-code-plugins/query
Clone the repo
git clone --depth 1 https://github.com/doneyli/claude-code-plugins
Per session 19 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 578 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00019 $0.00578
Opus 5 $0.00010 $0.00289
Sonnet 5 $0.00004 $0.00116
Haiku 4.5 $0.00002 $0.00058

Measured yesterday against content hash 8fb34b6542d1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

query 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 yesterday.

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.

llm-wiki/commands/query.md · 66 lines

What it actually says

Answer a question grounded in the wiki at the current working directory.

Parse $ARGUMENTS:

  • The quoted string is the question.
  • If --file-back appears, write the answer back as a synthesis page.

Steps:

  1. Read wiki/index.md to orient.
  2. Identify the 5-20 most relevant pages based on the question. Prefer:
    • Higher confidence in frontmatter
    • Fresher last_updated
    • Breadth over depth (10 medium-relevance pages beats 3 highly-relevant ones)
  3. Read each identified page in full.
  4. Synthesize an answer. Every factual claim cites a wiki page via [[wikilink]].
  5. Flag gaps with > [!gap] if the wiki lacks coverage the question needs. Flag contradictions with > [!contradiction] if relevant pages disagree.

If --file-back was passed:

  1. Create wiki/synthesis/<slug>.md where slug is a kebab-cased version of the question (max 60 chars). Use frontmatter:
    ---
    title: <Question>
    type: synthesis
    tags: [query-output]
    sources: [<list of wiki pages drawn on>]
    created: YYYY-MM-DD
    last_updated: YYYY-MM-DD
    ttl: 90d
    confidence: <lowest confidence among sources>
    ---
    
  2. Write the answer body below the frontmatter. End with a ## Backlinks section (empty if this is the first reference).
  3. Link the new synthesis page from wiki/index.md under ## Synthesis.

Always:

  1. Append a query log entry to wiki/log.md:
    ## [YYYY-MM-DD] query | <question>
    - Pages consulted: [[a]], [[b]]
    - Answer filed at: wiki/synthesis/<slug>.md   (or "not filed")
    - Gaps identified: [...]
    

Quality bar:

  • Do not invent facts not present in the pages you read.
  • If the wiki has insufficient coverage, say so. Do not extrapolate.
  • Confidence in your answer is bounded by the lowest-confidence source page.
  • If pages have expired TTLs, note: "This answer draws on pages that may be stale."

Print the answer directly. Follow with:

Sources: [[page-a]], [[page-b]], ...
Confidence: high | medium | low
Gaps: [...] or "none identified"
Filed back: wiki/synthesis/<slug>.md or "not filed"
Changes

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.

  1. yesterday First seen · 66 lines · 19 tokens per session scan A 8fb34b6542d1

Subscribe to this mod's changes

query is a command published in the GitHub repository doneyli/claude-code-plugins (4 stars, last pushed 4mo ago), licensed MIT. It adds 19 tokens to every session and 578 once invoked, about $0.0001 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.