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 zjunlp/Mechanist --skill mechanic-db-searchgit clone --depth 1 https://github.com/zjunlp/MechanistWrote 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/zjunlp/mechanist/mechanic-db-search)<a href="https://agentmods.dev/skills/zjunlp/mechanist/mechanic-db-search"><img src="https://agentmods.dev/badge/skills/zjunlp/mechanist/mechanic-db-search/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/zjunlp/mechanist/mechanic-db-search"><img src="https://agentmods.dev/badge/skills/zjunlp/mechanist/mechanic-db-search.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 | $0.00057 | $0.08297 |
| Opus 5 | $0.00028 | $0.04149 |
| Sonnet 5 | $0.00011 | $0.01659 |
| Haiku 4.5 | $0.00006 | $0.00830 |
Grade A, and why
mechanic-db-search 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 11d 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 — 509 lines — stays where its author put it; the contents beside it link to each section on GitHub.
mechanic-db-search — Cloud SEARCH Wrapper
Query: $ARGUMENTS
What this skill does
Wrap the cloud-hosted mechanic_database SEARCH service so the Agent can:
- Submit a search — preferably as an Agent-built decomposed query JSON that encodes at most one sub-query per database (db choice, closed-enum interp fields, keywords, and an Agent-written HyDE abstract per sub-query).
- Receive a paper list (
papers[]) and use it downstream. - Iterate: if results are unsatisfactory, refine and call again (up to 3 rounds).
- Run multiple calls in parallel when appropriate.
All transport — auth, submit, poll, error handling, writing results to disk — lives in the mechanic-db MCP server (mcp-servers/mechanic-db/server.py), which exposes a single tool, search_papers. The MCP server only forwards the request: it does not validate the decomposition or run any splitter/HyDE itself — query correctness is this skill's job. The heavy retrieval — multi-ranker scoring + fusion, and (in flat mode) the splitter and HyDE generation — runs in the cloud SEARCH service. Call the search_papers tool directly; do not invoke any Python.
Constants
- TOOL = the
search_paperstool on themechanic-dbMCP server. Call it directly; do not invoke any Python. - API_KEY — provided to the MCP server via the plugin manifest env mapping (
MECHANIC_DB_API_KEY; set as an environment variable in the shell that launches Claude Code, or for Codex the cachedmcp.codex.jsonenv). Optional: with no key the service still answers on an anonymous per-IP tier (2 searches/min, 20/day, vs 20/min and 1000/day with a key). An unset key is therefore not a reason to skip this skill. - DEFAULT_TOP_K = 300
- TIMEOUT_SEC = 1200 (one call is typically 3-7 min; allow headroom)
- MAX_REFINEMENT_ROUNDS = 3
Skip-when-unconfigured contract
What the service searches
| ID | Scope |
|---|---|
interp_db |
AI interpretability papers (~14k) — neural-network internal mechanisms in LLMs / Transformers / CNNs. Tagged along closed-enum axes (techniques / task_scenarios / abilities / components) plus free-text target_models / model_families. |
sciatlas_db |
All-discipline graph (~157M nodes) — neuroscience, psychology, cognitive science, computer science, biology, physics, chemistry, materials, humanities. Retrieved via semantic similarity + BM25. |
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.
- 11d ago First seen · 509 lines · 57 tokens per session scan A 0b4c264ffea4
mechanic-db-search is a skill published in the GitHub repository zjunlp/Mechanist (74 stars, last pushed 15d ago), licensed MIT. It adds 57 tokens to every session and 8,297 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
nanoresearch-writing
Draft a LaTeX research paper from all previous stage outputs.
nanoresearch-experiment
Generate a Python code skeleton from an experiment blueprint.
nanoresearch-planning
Produce an experiment blueprint from a research hypothesis.
nanoresearch-ideation
Search academic literature and generate research hypotheses.
autoresearch
Orchestrates end-to-end autonomous AI research projects using a two-loop architecture. The inner loop runs rapid experiment iterations with clear optimization targets. The outer loop synthesizes results, identifies patterns, and steers research direction. Routes to domain-specific skills for execution, supports…
skypilot-multi-cloud-orchestration
Multi-cloud orchestration for ML workloads with automatic cost optimization. Use when you need to run training or batch jobs across multiple clouds, leverage spot instances with auto-recovery, or optimize GPU costs across providers.