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 mehdiforoozandeh/crux --skill crux-cockpitgit clone --depth 1 https://github.com/mehdiforoozandeh/cruxWrote 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/mehdiforoozandeh/crux/crux-cockpit)<a href="https://agentmods.dev/skills/mehdiforoozandeh/crux/crux-cockpit"><img src="https://agentmods.dev/badge/skills/mehdiforoozandeh/crux/crux-cockpit/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/mehdiforoozandeh/crux/crux-cockpit"><img src="https://agentmods.dev/badge/skills/mehdiforoozandeh/crux/crux-cockpit.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 | $0.00197 | $0.04630 |
| Opus 5 | $0.00098 | $0.02315 |
| Sonnet 5 | $0.00039 | $0.00926 |
| Haiku 4.5 | $0.00020 | $0.00463 |
Grade C, and why
crux-cockpit scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
curl -fsS "$URL/snapshot.json" | python3 -c \ Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
and `/snapshot.json` must have answered 200 from *your* curl. A URL you haven't verified How it starts
The opening of the file, as written. The whole thing — 291 lines — stays where its author put it; the contents beside it link to each section on GitHub.
crux-cockpit — launch the crux GUI, beginning to finish
The cockpit is crux's read-only browser GUI: crux serve boots a stdlib HTTP server
on 127.0.0.1, serves the webui/ frontend, and exposes the vault as /snapshot.json
(status-colored tree — pan / zoom / collapse / search / re-orient / focus one question —
plus the review queue, node detail with rendered markdown, and a hypothesis's linked report
and figures; live-refreshing, zero writes). Either panel can go full-screen ([ / ],
Esc restores). The engine does the serving; you (the
agent) do the operating: find the vault, launch clean, verify, hand over one URL, and
manage the server's lifecycle. The user should never have to touch a terminal on the
machine you run on — the one thing you can't do for them is start a local ssh -L
tunnel (plain-SSH contexts only; see Local vs remote).
The engine rides in the crux skill: ENGINE = <crux skill>/scaffold/crux.py — the
crux skill is this skill's sibling directory in your skills dir (in a repo clone,
skills/crux). If that engine file is missing, the crux skill isn't installed: install
it first (npx skills add mehdiforoozandeh/crux --all, or the repo's ./install.sh)
before continuing.
Ground rules (load-bearing)
- Never report an unverified URL. Before telling the user the cockpit is up, both
/and/snapshot.jsonmust have answered 200 from your curl. A URL you haven't verified does not leave your mouth. - URL only — never pop a browser. Always launch with
--no-openand deliver the URL as a clickable link. The user opens it when ready. One carve-out: in VS Code Remote-SSH the sanctioned way to create the port forward is an open call (code --openExternal— see the ladder below); there, opening the user's browser is the mechanism, not a discourtesy. - Always fresh, per vault. A launch first kills any cockpit already serving that vault — and never touches cockpits serving other vaults (several can coexist on different ports).
- The cockpit is read-only. No route writes; every mutation stays in the agent/CLI. Never present it as an editor.
- Surface engine-drift warnings. If the launch log warns that the vault was stamped by a different engine version, relay that warning to the user verbatim.
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.
- 12d ago First seen · 291 lines · 197 tokens per session scan C ab41eb762337
crux-cockpit is a skill published in the GitHub repository mehdiforoozandeh/crux (6 stars, last pushed 6d ago), licensed MIT. It adds 197 tokens to every session and 4,630 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, 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
ab-testing
Use when designing or analyzing a controlled experiment — falsifiable hypothesis, sample size from an MDE, reading significance/CI/power, CUPED, or rescuing tests that won't go significant. NOT recurring metric tracking (that is analytics), NOT north-star/KPI trees (that is kpi-framework), NOT projecting metrics…
product-principles
Defines 4 Risks confidence thresholds, OST hierarchy levels, Knowledge Pyramid tiers, and state design requirements. Use when evaluating user stories, setting confidence scores, referencing OST levels, scoping MVP, or determining validation sufficiency.
recipe-define
Orchestrate PRD creation from validated hypotheses — standard PRD output with 4 Risks confidence and hypothesis traceability.
recipe-validate
Orchestrate hypothesis validation through type-appropriate methods — prototypes, code analysis, market research, and expert review.
hypothesis-discipline
Manages hypothesis lifecycle, enforces validation criteria, time budgets, and confidence scoring rules. Use when creating hypotheses, updating confidence scores, setting validation criteria, handling timeouts, or recording validation results.
blueprint-standards
Defines structural design artifact formats — information architecture, user flows, content model, brand direction, and AI interaction model. Use when creating or reviewing structural design documents that precede prototype generation.