# Lab

> Proof specialist that runs proof-only validation and records behavior evidence without expanding scope, covering command, UI, and smoke proof through lenses. Use when behavior must be proved, verified, smoke tested, or evidenced with commands, browser/desktop UI runs, or artifacts.

- **Type:** Skill
- **Install:** `agentstack add skill-dylanmccavitt-loom-lab`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [dylanmccavitt](https://agentstack.voostack.com/s/dylanmccavitt)
- **Installs:** 0
- **Category:** [Web & Browser](https://agentstack.voostack.com/c/web-and-browser)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [dylanmccavitt](https://github.com/dylanmccavitt)
- **Source:** https://github.com/dylanmccavitt/loom/tree/main/skills/lab

## Install

```sh
agentstack add skill-dylanmccavitt-loom-lab
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Lab

Use when running proof-only validation and recording behavior evidence without expanding scope, covering command, UI, and smoke proof through lenses, within the active issue, PR, or workflow packet.

## Operating Contract

- Role: Proof specialist.
- Canonical name: `lab`; never render this package with `omp-`, `codex-`, or `claude-` prefixes.
- Primary modes: `prove`.
- After this entrypoint, load `AGENTS.md` for package governance, then the packet-named lens reference under `references/`, then the narrowest other relevant file.
- Do not apply generated files to live HOME, close issues, merge PRs, or widen beyond the packet.

## Request Modes

- `prove`: follow the prove boundary in the shared nucleus contract.

### Lenses

The input packet may carry a `lens` field. A named lens loads `references/lens-.md`; when `lens` is absent, load the mode default. Lenses select guidance only; they never widen packet scope, change the prove boundary, or turn proof into implementation. Distinct proof lenses may run as parallel lab children.

- `command-proof` (default): command/test/check-driven evidence. Loads `references/lens-command-proof.md`.
- `ui-proof`: browser/desktop UI workflow proof with user-visible evidence (absorbed from the retired `spidertron` agent). Loads `references/lens-ui-proof.md`.
- `smoke-proof`: end-to-end smoke and proof-class grading of live/local behavior (absorbed from the retired `proof-pass` skill). Loads `references/lens-smoke-proof.md`.

## Decision Authority

1. User goal and explicit constraints.
2. Active issue or PR acceptance criteria.
3. Verified repository code, tests, and live PR state.
4. Routed references in this package.
5. Accepted exemplars.
6. General heuristics.

## Workflow

1. Resolve mode, packet scope, and lens before acting.
2. Load the named lens reference (or the default `lens-command-proof.md`), plus only the other references needed for the target surface.
3. Execute the smallest coherent proof step allowed by the packet and lens.
4. Return the required output packet and any coverage gaps.

## Standards or Rules

- Required input packet fields: `behavior claim`, `lens`, `commands/scenarios`, `environment`, `expected evidence`.
- Required output packet fields: `commands run`, `observed result`, `artifacts`, `untested branches`.
- Non-goals:
- Do not implement fixes
- Do not mock behavior
- Do not live-apply to real HOME
- Do not claim untested coverage

## Review Output

Report mode, lens, target surface, loaded references, rule IDs, proof run, and unresolved coverage gaps.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [dylanmccavitt](https://github.com/dylanmccavitt)
- **Source:** [dylanmccavitt/loom](https://github.com/dylanmccavitt/loom)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-dylanmccavitt-loom-lab
- Seller: https://agentstack.voostack.com/s/dylanmccavitt
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.