Borrowing it
Nothing to install: this file belongs to supertrained/rhumb. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/supertrained/rhumb/main/.cursor/skills/verify-rhumb/SKILL.mdgit clone --depth 1 https://github.com/supertrained/rhumbWrote 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/supertrained/rhumb/verify-rhumb)<a href="https://agentmods.dev/skills/supertrained/rhumb/verify-rhumb"><img src="https://agentmods.dev/badge/skills/supertrained/rhumb/verify-rhumb/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/supertrained/rhumb/verify-rhumb"><img src="https://agentmods.dev/badge/skills/supertrained/rhumb/verify-rhumb.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 | $0.00061 | $0.01910 |
| Opus 5.5 | $0.00024 | $0.00764 |
| Sonnet 5.5 | $0.00012 | $0.00382 |
| Haiku 4.5 | $0.00006 | $0.00191 |
Grade A, and why
verify-rhumb scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
Primary surface is the public HTTP API. MCP, CLI, and the website wrap the same Index and Resolve reads. This skill drives those reads with curl. It does not call `POST /v1/capabilities/{id}/execute` or any paid rail. How it starts
The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verify Rhumb
Read this file, then features/README.md, then the one feature file you will drive. Do not invent capabilities. Do not execute.
Primary surface is the public HTTP API. MCP, CLI, and the website wrap the same Index and Resolve reads. This skill drives those reads with curl. It does not call POST /v1/capabilities/{id}/execute or any paid rail.
Launch
Default target is live production. Do not start a local process.
export VERIFY_RHUMB_BASE="${VERIFY_RHUMB_BASE:-https://api.rhumb.dev}"
export VERIFY_RHUMB_SITE="${VERIFY_RHUMB_SITE:-https://rhumb.dev}"
export VERIFY_RHUMB_RUN_ID="${VERIFY_RHUMB_RUN_ID:-$(date -u +%Y%m%dT%H%M%SZ)}"
.cursor/skills/verify-rhumb/bin/launch
Ready means GET $VERIFY_RHUMB_BASE/v1/healthz returns HTTP 200 and {"status":"ok"}.
Teardown is bin/cleanup. Live launch creates no process. Cleanup must still run so scratch state under /tmp/verify-rhumb-state-$VERIFY_RHUMB_RUN_ID is removed.
Optional local API
This checkout documents local start in the Makefile and packages/api/README.md.
make dev
cd packages/api && python3.12 -m venv .venv && source .venv/bin/activate
pip install -r requirements-dev.txt
uvicorn main:app --reload --port 8000
packages/api/main.py re-exports app from app.py. The root README uvicorn app:app line also works. Prefer uvicorn main:app --port 8000.
Local search and resolve read Supabase through packages/api/routes/_supabase.py. .env.example sets SUPABASE_URL=http://localhost:54321 and placeholder keys. Without a real Supabase catalog, /v1/search returns empty results or "Search unavailable." That is not a production proof.
make dev needs Docker for Postgres 54322 and Redis 6379. If docker is missing, stay on https://api.rhumb.dev.
Astro web, if you need it: cd packages/astro-web && npm ci && npm run dev on port 4321. MCP discovery: npx -y --package rhumb-mcp@latest rhumb-mcp with no key.
Doctor
Run this first whenever the instance looks off.
What ships with it
24 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.
- bin/cleanup 862 B
- bin/common.sh 1.6 KB runs code
- bin/doctor 2.2 KB
- bin/drive 9.8 KB
- bin/launch 1.9 KB
- bin/prove 1.2 KB
- evidence/20260910T014200Z/doctor.json 1.9 KB
- evidence/20260910T014200Z/drive-index-search.json 364 B
- evidence/20260910T014200Z/http/doctor-healthz.json 15 B
- evidence/20260910T014200Z/http/doctor-healthz.meta 588 B
- evidence/20260910T014200Z/http/doctor-status.json 1.0 KB
- evidence/20260910T014200Z/http/doctor-status.meta 612 B
- evidence/20260910T014200Z/http/healthz.json 15 B
- evidence/20260910T014200Z/http/healthz.meta 588 B
- evidence/20260910T014200Z/http/search-email.json 2.5 KB
- evidence/20260910T014200Z/http/search-email.meta 628 B
- evidence/20260910T014200Z/launch.json 280 B
- evidence/README.md 264 B
- features/capability-discover.md 1.9 KB
- features/index-score.md 2.1 KB
- features/index-search.md 3.4 KB
- features/README.md 2.8 KB
- features/resolve-search-query.md 2.7 KB
- features/typed-stop-unknown-capability.md 2.6 KB
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 · 163 lines · 61 tokens per session scan A a2bddee3cf34
verify-rhumb is a skill published in the GitHub repository supertrained/rhumb (5 stars, last pushed 16d ago), licensed MIT. It adds 61 tokens to every session and 1,910 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-11.
Other skills, from other repositories
mandu-testing
Testing patterns for Mandu applications. Use when writing unit tests, integration tests, or E2E tests. Triggers on test, spec, Bun test, Playwright, or testing tasks.
api-test
Run API integration tests against the running backend, verify endpoints return expected responses and status codes. Use after deploying a preview or starting the dev server.
verify-omnigent
Drive Omnigent the way a user does and prove a behavior with recorded evidence, using an isolated server, runner, and mock model plus a feature map of every user entry point. Load before reproducing a user-facing bug, before claiming a fix works, or when reviewing whether a change covered every surface (session vs.…
test-master
Generates test files, creates mocking strategies, analyzes code coverage, designs test architectures, and produces test plans and defect reports across functional, performance, and security testing disciplines. Use when writing unit tests, integration tests, or E2E tests; creating test strategies or automation…
eval-dataset-design
A guide to creating reliable test sets for AI agents. Each evaluation task includes the request, the simulated user's behaviour, a reset starting state, and a success check that can be independently verified.
dogfood
Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to dogfood, QA, exploratory test, find issues, bug hunt, or test this app on mobile.