Borrowing it
Nothing to install: this file belongs to duyet/oma. 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/duyet/oma/main/.cursor/skills/verify-oma/SKILL.mdgit clone --depth 1 https://github.com/duyet/omaWrote 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/duyet/oma/verify-oma)<a href="https://agentmods.dev/skills/duyet/oma/verify-oma"><img src="https://agentmods.dev/badge/skills/duyet/oma/verify-oma/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/duyet/oma/verify-oma"><img src="https://agentmods.dev/badge/skills/duyet/oma/verify-oma.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 | $0.00055 | $0.01775 |
| Opus 5.5 | $0.00022 | $0.00710 |
| Sonnet 5.5 | $0.00011 | $0.00355 |
| Haiku 4.5 | $0.00006 | $0.00178 |
Grade A, and why
verify-oma 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 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.
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 — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verify oma
Agent-facing control skill for this repo. Launch a disposable instance, doctor it, drive a mapped feature, capture evidence, then clean up only what this run started.
Primary surface: the marketing site at apps/web (Astro, no secrets). Secondary surface: the operator Console at apps/console (Vite on :5173, proxies /v1 and /auth to :8787). The API worker (pnpm dev, :8787) is out of scope for a no-secrets run. Do not invent BETTER_AUTH_SECRET, PLATFORM_ROOT_SECRET, or API keys.
The lever is control-oma.mjs in this directory. Prefer it over throwaway scripts.
node .cursor/skills/verify-oma/control-oma.mjs <command>
Repo root is detected from the skill file. Override with OMA_ROOT. Evidence lives under .cursor/skills/verify-oma/artifacts/<run-id>/ and is copied to /opt/cursor/artifacts/ when that directory exists. Cleanup never deletes evidence.
Launch
One surface per run. Two instances may share a machine if they use different ports. Never attach to a server this run did not start.
Marketing (default, no secrets):
node .cursor/skills/verify-oma/control-oma.mjs launch web
Starts pnpm exec astro dev --host 127.0.0.1 --port <port> from apps/web. Default port 4321. Ready when GET / returns HTTP 200 and the body contains Open Managed Agents. Logged URL is stored in .cursor/skills/verify-oma/.run/state.json.
Console (UI only; API optional):
node .cursor/skills/verify-oma/control-oma.mjs launch console
Starts pnpm exec vite --host 127.0.0.1 --port <port> --strictPort from apps/console. Default port 5173. Ready when GET /login returns HTTP 200. Does not start the worker. Login can render without a live API. Protected routes (/, /agents, /analytics) redirect to /login when unauthenticated (AppShell).
Pass --port N to either launch. Isolation: pick a free port. Do not reuse 4321/5173 if they already answer.
Teardown is cleanup, not killall.
What ships with it
11 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.
- .gitignore 17 B
- control-oma.mjs 25 KB runs code
- features/console-agents.md 1.6 KB
- features/console-analytics.md 1.9 KB
- features/console-inject.md 1.9 KB
- features/console-login-mobile.md 1.9 KB
- features/console-login.md 2.0 KB
- features/console-session-artifacts.md 1.9 KB
- features/landing-features.md 1.7 KB
- features/landing-home.md 2.2 KB
- features/README.md 3.6 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 Changed 2be45a05d2be
- 1mo ago Changed c81f9fc0898f
- 1mo ago First seen · 151 lines · 55 tokens per session scan A da16b4be261c
verify-oma is a skill published in the GitHub repository duyet/oma (10 stars, last pushed 9d ago), licensed Apache-2.0. It adds 55 tokens to every session and 1,775 once invoked, about $0.0002 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 2026-09-04.
Other skills, from other repositories
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.
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…
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.
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
critique
Adversarial design critique of the current work. Render it, look at it, and argue for rejection. Run after the gates are green, never instead of them.