agentleFS
Sign inStart free
frankxaifrankxai/starlight-agent-skills/skills/studios/software-studio/SKILL.md

software-studio

Skill1 starsChanged 4 months agoMITView on GitHub

Build and release secure applications, mobile products, games, plugins, MCP servers, agents, SDKs and platform modules with professional SDLC evidence. Use when a software product needs bounded architecture, implementation, adversarial review, exact-revision verification, deployment packaging or marketplace preparation.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run format, lint, type, unit, integration, permission, retry, malformed-input, dependency, secret and migration gates proportional to risk.

    line 31
  2. Very specific:

    Verify accessibility, responsive behavior, performance, compatibility, empty/loading/error/recovery paths, observability, model/runtime …

    line 32

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Software Studio

Purpose

Turn a product claim into a bounded, secure, observable and maintainable software release while preserving portable domain contracts and exact operational evidence.

When it fires

  • Building or materially changing an application, mobile app, game, plugin, MCP server, agent, SDK or platform module.
  • Preparing a Vercel, Railway, Apple App Store, Google Play, itch.io or owned-product release.
  • Repairing a failing, stale, unlinked or unverifiable software release.

Inputs

  • Product manifest, repository instructions, exact base revision and accepted architecture decisions.
  • Software Studio task envelope with scope, non-goals, token/cost ceilings, permissions, target channels, acceptance, rollout and stop conditions.
  • Threat context, data/identity/tool contracts, supported-version matrix and existing runtime evidence.

Workflow

  1. Resolve the canonical product repository, Software Studio pack, architecture decisions and channel adapters. Do not create another platform, service or repository from naming convenience.
  2. Define the measurable user job, critical path, non-goals, interfaces, state, failure behavior, security boundary, telemetry and release evidence before implementation.
  3. Assign product architecture, implementation, security/reliability review, product-quality review and exact-revision verification to non-overlapping owners.
  4. Work on a reviewable branch. Preserve unrelated changes; keep secrets out of source; use reversible migrations, idempotency and explicit rollback.
  5. Run format, lint, type, unit, integration, permission, retry, malformed-input, dependency, secret and migration gates proportional to risk.
  6. Verify accessibility, responsive behavior, performance, compatibility, empty/loading/error/recovery paths, observability, model/runtime cost and support boundaries.
  7. Build a preview or installable package from the exact revision. Compile requested store, template or integration packages using the registered adapters.
  8. Record tests, independent review, preview/deployment, smoke evidence, known limitations, rollback and human-governed actions in the release receipt.

Output contract

Return architecture decision, reviewable revision, test/security/quality evidence, preview or installable artifact, channel packages, SLO/cost/rollback record, exact-revision release receipt and release decision.

Tools and MCP

Use GitHub for source, review and CI; Vercel for Next.js experiences and application-local jobs; Cloudflare for accepted persistent entity or edge boundaries; Supabase/Postgres for shared business truth; Linear for execution projection; and browser/runtime tools for observable smoke evidence. MCP is a typed boundary, not product authority.

Quality bar

  • Architecture, permissions, data ownership, failure modes and recovery are explicit.
  • Required gates pass on the exact release revision.
  • Preview, production and public availability are reported as different states.
  • Operability, observability, cost, rollback, support and compatibility are owned.
  • Merge, production promotion, store submission, pricing, spending and external contact follow their authorization gates.

Built on SIP — Starlight Intelligence Protocol Vertical: federated-product-studios · portable capability layer

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free