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 agentmods add commands/datathings/marketplace/docsgit clone --depth 1 https://github.com/datathings/marketplaceWhat 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 | $0.00012 | $0.02877 |
| Opus 5 | $0.00006 | $0.01438 |
| Sonnet 5 | $0.00002 | $0.00575 |
| Haiku 4.5 | $0.00001 | $0.00288 |
Grade D, and why
docs scanned grade D with 3 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 2d 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.
Sends data to an external URLmediumData exfiltration
A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.
curl -H "task: true" -X POST -d '[]' http://localhost:8080/<module>::<fn> # returns task_id Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
11. **Database** — dev reset (`rm -rf gcdata && greycat run import`); backup/restore via `greycat backup` / `greycat restore <archive>` (prefer over raw `tar` on a live `gcdata/`) Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
await axios.post('/<module>::<fn>', [param1, param2]); How it starts
The opening of the file, as written. The whole thing — 223 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Documentation Generator
Purpose: Generate/update README.md, API.md, MCP.md (+ optional OpenAPI, frontend SEO artifacts).
Run after: each sprint, before releases, when adding features/APIs.
Regenerate after: new endpoints, model changes, lib upgrades, auth changes, new routes (refresh sitemap.xml/llms.txt). Not for bug fixes, internal refactors, or doc-only changes.
ultrathink + ultracode — read the actual code, never emit boilerplate. README.md (feature scan), API.md (@expose scan), MCP.md (@tag("mcp") scan), and the Phase-5 SEO/llms.txt artifacts are INDEPENDENT deliverables: when parallel subagents / ultracode are available, draft them as PARALLEL subagents (one each), then assemble & write. Otherwise generate sequentially.
Frontend stack (referred to below as the frontend stack): VitePlus (vp) + Lit (shadow DOM) + TypeScript + Web Awesome (or equivalent) + @greycat/web/sdk + lucide-static, pnpm; note i18next / maplibre-gl / Vitest (optional) if present.
Lighthouse (referred to below): audit both mobile (default) and desktop (--preset=desktop), target ≥ 90 per category — via the project's lighthouse script (:desktop / :ci variants) if present, else the lighthouse CLI against the served app (greycat serve/dev first). Optional tooling.
Phase 1: README.md
Detect features
# Frontend: frontend/ directory, or lit / @awesome.me/webawesome / @greycat/web in package.json
# Tests: test/ directory
# Libs: @library in project.gcl
# Auth: @permission / @role in project.gcl
# MCP: @tag("mcp") in src/
# Data: data/ directory, *.gguf files
# SEO: webroot/{robots.txt,sitemap.xml,llms.txt,*.webmanifest}
Extract info
grep "@library" project.gcl # versions
grep "@permission\|@role" project.gcl # auth
grep -rn "^type [A-Z]" src/ --include="*.gcl" # models
grep -rnE "^abstract type.*Service|^(static )?fn [a-z_]" src/ --include="*.gcl" # services (abstract-type OR free fns)
grep -rn "@expose" src/ --include="*.gcl" # endpoints (@expose may live in src/api.gcl, not only *_api.gcl)
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.
- 2d ago First seen · 223 lines · 12 tokens per session scan D 04d9b2f0af71
docs is a command published in the GitHub repository datathings/marketplace (11 stars, last pushed 5d ago), licensed Apache-2.0. It adds 12 tokens to every session and 2,877 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it D with 3 findings (sends data to an external url, recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.