launchplane

Inspect and mutate Launchplane-managed product and runtime state through audited helper scripts.

5|1|Updated May 5, 2026
One-click install
npx skills add https://github.com/cbusillo/codex-skills --skill launchplane-cbusillo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: launchplane
Source: https://github.com/cbusillo/codex-skills/tree/main/skills/launchplane
Command: npx skills add https://github.com/cbusillo/codex-skills --skill launchplane-cbusillo

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? Managing Launchplane-controlled product, runtime, deployment, and merge-train state requires strict authority boundaries, redacted output, and reviewed dry-run/apply discipline that raw API calls and ad-hoc scripts cannot guarantee. ## Core Features & Use Cases - Context Reading: Read public-safe Launchplane context for a repository, branch, issue, or pull request via the launchplane-context helper with fail-closed redaction. - Bounded Write Actions: Preflight, dry-run, and apply product-config, repository-inventory, change-impact policy, merge-train policy, and deploy-recovery changes with idempotency keys and digest-bound reviewed evidence. - Contract Governance: Validate the vendored agent/operator contract offline and compare its semantic freshness against the upstream artifact. - Use Case: An operator needs to import a new merge-train policy. The skill dry-runs the private payload after active-policy digest preflight, then applies only the exact reviewed evidence with an idempotency key, emitting redacted output. ## Quick Start Ask the agent to read Launchplane context for your repository with the launchplane-context helper, then dry-run any intended product-config or policy change before applying it.

Frequently Asked Questions about launchplane

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I read Launchplane context for a repository?

Run the launchplane-context helper with the repository in OWNER/REPO form, optionally adding branch, issue, or pull request numbers. It emits public-safe JSON with a status of available, no_context, unavailable, unauthorized, or invalid, and skills continue normally on any non-available status.

How do I apply a Launchplane product-config change safely?

Use the write-action helper's product-config-preflight, product-config-dry-run, then product-config-apply sequence with a payload file and idempotency key. Raw curl or HTTP calls to the product-config API are blocked by command policy because they bypass redaction and dry-run discipline.

Can I call the Launchplane API directly with curl?

No. Command policies require the bundled write-action helper for merge-train, product-config, change-impact policy, repository-inventory, and deploy-recovery routes. Direct HTTP calls bypass private-file handling, reviewed evidence, idempotency, and redaction.

Does the contract validation check upstream freshness?

The offline conformance gate only proves the vendored contract and local consumers are internally consistent. Use the freshness script's compare command for advisory current, known-stale, or unknown semantic-drift evidence against the upstream artifact.

Why are GitHub Actions secret writes rejected for Launchplane?

Writing Launchplane authorization state through GitHub secrets or variables would silently make GitHub an authorization authority. The skill rejects those writes and routes authorization gaps to the DB-native authorization architecture instead.

When should I not use checked-in repo metadata for runtime values?

Never treat checked-in files, workflow variables, or archived workstation configs as live runtime authority. Launchplane service records or explicit operator input own product, tenant, domain, lane, and health-check values; repo metadata is routing context only.