Skip to content
Back to skills

Autopilot Lab

ASecurity

Use when experiments need setup, evaluation/analysis, failure reproduction, or result reporting, including fixed results without new measurements. Owns experiments; default preset. Not for reusable evaluation code, independent documents, or fixed-data document layout/caption edits.

  • 8 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 3, 2026
ai-agents

Security analysis

A100/100

Scanned October 5, 2026

npx -y skills add dmlguq456/hearting --skill autopilot-lab --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Autopilot Lab?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Autopilot Lab
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/dmlguq456-autopilot-lab-856a554f/badge)](https://www.skillsdirectory.com/skills/dmlguq456-autopilot-lab-856a554f)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: autopilot-lab
description: "Use when experiments need setup, evaluation/analysis, failure reproduction, or result reporting, including fixed results without new measurements. Owns experiments; default preset. Not for reusable evaluation code, independent documents, or fixed-data document layout/caption edits."
metadata:
  portable_source: capabilities/autopilot-lab.md
  adapter: opencode
  invocation_class: entry-router
---

# autopilot-lab

This is an OpenCode-native Skill projection generated from the portable
capability contract. It is adapter-owned output, not a legacy compatibility Skill copy.

## Source

- Portable source: `capabilities/autopilot-lab.md`
- Runtime check: `adapters/opencode/bin/preflight.sh capability-info autopilot-lab`
- Bootstrap: `adapters/opencode/AGENTS.md`

## Use

1. Before approval, route from this compact metadata and `core/WORKFLOW.md §0.2`; do not read the full portable source merely to propose the route.
2. Present the five-field confirmation card from `core/WORKFLOW.md §0.4` unless the same route and scope are already approved.
3. After approval, direct/quick acting sessions read `capabilities/autopilot-lab.md`; at `standard+`, the dispatch-depth-1 owner reads it and stage workers read only their assigned contracts.
4. Before the first durable capability artifact, compile and bind the checked route with `preflight.sh route --capability autopilot-lab ...`. A native-agent restriction never authorizes direct execution.
5. Run `adapters/opencode/bin/preflight.sh capability-info autopilot-lab` and obey the reported status:
   - `instruction-only`: use this Skill as OpenCode guidance plus explicit preflight guards.
   - `tool-contract`: report the named `tool_contract`, run any `tool_contract_check`, and obey `runtime_surface` / `fallback` before claiming full support.
   - `unsupported`: stop or use the reported `fallback`.

## Shape

- Identifier: `autopilot-lab`
- Invocation class: `entry-router`
- Supported modes: `setup, eval`
- Argument shape: `<task description> [--mode setup|eval|auto] [--parent <slug>] [--ref <similar-model-path>] [--intensity direct|quick|standard|strong|thorough|adversarial] [--report] [--from spec|scaffold|run|eval|summary]`
- Portable meaning: Set up experiments, evaluate, and report results, including fixed results.



## Workflow Evidence

- Before durable capability output: `adapters/opencode/bin/preflight.sh route --capability autopilot-lab <complete compile arguments>`
- For workflow state: `adapters/opencode/bin/preflight.sh status [cwd] [session-id]` and `adapters/opencode/bin/preflight.sh prompt-signal [cwd] [session-id]`

Do not use legacy compatibility Skill files or non-native adapter Skill files
as OpenCode-native source. Those files are compatibility/reference surfaces only.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.