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 lestrrat-ai/claude-code-plugins --skill query-apigit clone --depth 1 https://github.com/lestrrat-ai/claude-code-pluginsWrote 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/lestrrat-ai/claude-code-plugins/query-api)<a href="https://agentmods.dev/skills/lestrrat-ai/claude-code-plugins/query-api"><img src="https://agentmods.dev/badge/skills/lestrrat-ai/claude-code-plugins/query-api/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/lestrrat-ai/claude-code-plugins/query-api"><img src="https://agentmods.dev/badge/skills/lestrrat-ai/claude-code-plugins/query-api.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00065 | $0.00940 |
| Opus 5.5 | $0.00026 | $0.00376 |
| Sonnet 5.5 | $0.00013 | $0.00188 |
| Haiku 4.5 | $0.00006 | $0.00094 |
Grade A, and why
query-api 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 2mo 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.
What it actually says
Query Fusion API
Invocation: Claude Code /fusion:query-api; Codex $fusion:query-api.
-
Resolve
scripts/query_fusion_api.pyfrom this activeSKILL.mdto an absolute path. NEVER use a plugin-root environment variable. -
Run one subcommand per question:
["python3", "<absolute-query_fusion_api.py>", "<subcommand>", "<argument>"]Question Subcommand Provenance and counts for the database in use infoFind names matching a term search <term>Full detail for a class, function, or one member show <name>All members of a class members <class>(--owndrops inherited ones)Base chain + direct subclasses tree <class>Which symbols' docs mention a term doc-search <term>showaccepts qualnames (adsk.fusion.ExtrudeFeatures), bare class names,Class.member, and bare member names; ambiguous input returns the candidate list — pick one and re-run.Member lookups include inherited members by default, and every one is reported against the class that declares it:
show Class.memberresolves a member declared on any base,show Classlists inherited members after the class's own, andmembers <class>groups them by declaring class. The bases are walked in the order Python itself resolves an attribute, so the class reported is the class the member really comes from.membersandtreetake a class; given anything else they report no such class. -
Database resolution order (explicit
--db, then user cache, then the bundled copy) is owned by the script's module docstring; pass--dbonly when the user names a database, and pass it before the subcommand — after it, it is a usage error and exits 2:["python3", "<absolute-query_fusion_api.py>", "--db", "<path>", "<subcommand>", "<argument>"]Database missing at every location → tell user to run
/fusion:compile-api(Codex$fusion:compile-api). -
Answer from script output only. Exit code 1 with a candidate list or "no matches" on stdout is a lookup miss, not a failure — refine the name and re-run, or report the miss. Exit code 1 with an
error:line on stderr is a real failure — relay it and stop. The causes are: no database; the file is not one; a database that opened but could not answer the subcommand (could not answer '<subcommand>'); and a class whose recorded bases have no consistent resolution order, which is refused rather than answered from a guessed order and names the class. The last two are defects in the database — report them as such, and recompile with/fusion:compile-api(Codex$fusion:compile-api).
The database tracks what the Fusion runtime defines, not what the reference stubs declare. The stubs
declare a few members the shipped module does not define, and compile-api drops those, so a member
this database does not list is one a call cannot reach — that is the answer to give, not a gap to
work around. info names what the database in use dropped (stub_only_members_dropped);
compile-api's STUB_ONLY_MEMBERS owns the list and the reasoning.
For several related lookups, start the optional local JSON Lines session with serve and keep the
child process for the questions. Its exact request and response schema, error behavior, and database
snapshot contract are documented in the query session protocol.
Read-only. The bundled database ships with the plugin; NEVER edit or regenerate it here — regeneration
is compile-api's job.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2mo ago Changed · +4 lines f5dc7a0947a9
- 2mo ago Changed · +6 lines 873e516336c2
- 2mo ago First seen · 57 lines · 65 tokens per session scan A fc6fb9a37671
query-api is a skill published in the GitHub repository lestrrat-ai/claude-code-plugins (20 stars, last pushed 1mo ago), licensed MIT. It adds 65 tokens to every session and 940 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 2026-08-30.
query-api compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| query-apilestrrat-ai/claude-code-plugins | Skill | 20 | 1mo ago | MIT | A | 65 | Claude Code |
| imagegenopenai/codex | Skill | 134k | today | Apache-2.0 | A | 113 | Codex |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | yesterday | MIT | A | 200 | Claude Code, Codex |
| playwright-climicrosoft/playwright | Skill | 100k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| browser-automationopenclaw/openclaw | Skill | 394k | today | MIT | A | 36 | Claude Code |
| systematic-debuggingobra/superpowers | Skill | 315k | 5d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 95 | Claude Code |