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 DanielBirk04/capusqa --skill capusqa-helpgit clone --depth 1 https://github.com/DanielBirk04/capusqaWrote 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/danielbirk04/capusqa/capusqa-help)<a href="https://agentmods.dev/skills/danielbirk04/capusqa/capusqa-help"><img src="https://agentmods.dev/badge/skills/danielbirk04/capusqa/capusqa-help/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/danielbirk04/capusqa/capusqa-help"><img src="https://agentmods.dev/badge/skills/danielbirk04/capusqa/capusqa-help.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 | $0.00088 | $0.01112 |
| Opus 5.5 | $0.00035 | $0.00445 |
| Sonnet 5.5 | $0.00018 | $0.00222 |
| Haiku 4.5 | $0.00009 | $0.00111 |
Grade B, and why
capusqa-help scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Strips warnings and disclaimersmediumAnti-refusal
Omitting safety caveats hides risk from the user and is a common jailbreak preamble.
one `$capusqa-*` skill they need. Route, don't lecture. In Codex these skills are Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- **daemon down / connection refused / `curl` fails** → it isn't running: How it starts
The opening of the file, as written. The whole thing — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
capusqa-help — the CapusQA menu
You are the navigator: mirror the dashboard's left nav and send the user to the
one $capusqa-* skill they need. Route, don't lecture. In Codex these skills are
invoked with $capusqa-<name> (or via /skills), and Codex also picks them up
implicitly from the task.
Mental model (one breath, only if they're new)
CapusQA is persona-driven LLM testing for web and macOS apps. The daemon
(capusqa serve) is the single source of truth — every run, session, issue and
screenshot lives in ~/.capusqa. This chat and the dashboard
(http://127.0.0.1:7777/) are two clients of it: anything you do here shows in
the dashboard instantly and persists.
If they already said what they want, name the skill (don't show a menu)
- start/launch a test, "test my app", "find bugs" →
$capusqa-test - "what's running", show/watch runs, a session, steer/stop →
$capusqa-runs - bugs/findings, triage, "what did it find", send to my coding tool →
$capusqa-issues - configure anything (modes, run defaults, handoff/coding tool, findings filter, personas, audiences, logins) →
$capusqa-settings - first-time wiring / "nothing works yet" →
$capusqa-setup - update CapusQA / new version / vision models →
$capusqa-update - permissions, daemon health, clicks do nothing →
$capusqa-doctor - open the dashboard / the web UI →
$capusqa-dashboard
If they have no specific intent, ask once (numbered)
Four dashboard areas, recommended first (Setup, Update and Doctor are also available — they can just say so):
- New test (Home) — configure & start a test →
$capusqa-test - Runs — watch live / past runs →
$capusqa-runs - Issues found — findings & fixes →
$capusqa-issues - Settings — configure CapusQA →
$capusqa-settings
Then tell them the one skill to run (e.g. $capusqa-test) — don't make them guess.
The full map
| Skill | = dashboard | Does |
|---|---|---|
$capusqa-test |
New test (Home) | Configure & start a test (app URL, what-to-test, mode) |
$capusqa-runs |
Runs | Watch live/past runs, sessions, traces; steer/stop |
$capusqa-issues |
Issues found | Findings by severity, mark fixed, re-test, send to your coding agent |
$capusqa-settings |
Settings | Test modes · Run defaults · Handoff mode · Findings · Personas · Audiences · Saved secrets · Role logins · Usage · Your plan |
$capusqa-setup |
— | First-run: permissions, vision models, browser, project prep |
$capusqa-update |
— | Update CapusQA + skills + vision models |
$capusqa-doctor |
Home → System status | Environment health + fixes |
$capusqa-dashboard |
the web UI | Open the dashboard in your browser |
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 · 69 lines · 88 tokens per session scan B 9c88cdc4c58b
capusqa-help is a skill published in the GitHub repository DanielBirk04/capusqa (0 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 88 tokens to every session and 1,112 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it B with 2 findings (strips warnings and disclaimers, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
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.