Skip to content

matt-bat/got-skills / skills/help

Explain the installed Agent Command Center skillpack when the user sends --help, including every active skill, activation marker, supported command, and concise behavior description. Also supplies the exact one-line startup hint required when the pack is first loaded into a conversation; do not activate for unrelated help requests.

Help

Startup Hint

When the pack is first loaded in a new or existing conversation, print this exact line once:

send --help to learn more about available tools in agent-command-center

Do not repeat it every turn. The host instruction file owns startup enforcement because a skill cannot independently observe installation or conversation lifecycle events. When the host can execute local adapters, use scripts/runtime_adapter.py and preserve its returned conversation state. Read Host Runtime Protocol before integrating or testing a host.

--help Workflow

When the user sends exact command --help:

  1. Read skills/skill-catalog.json when it is available. Treat it as the canonical inventory and routing source.
  2. List every skill whose status is active. Include deprecated compatibility names in a separate note only when useful.
  3. Prefix a skill name with * exactly when its catalog routing_mode is explicit_only.
  4. Do not prefix automatic or safety-routed skills. Explain that they trigger when their typed conditions apply.
  5. For each skill, use the catalog description or a faithful one-line condensation.
  6. Show supported user commands and mode persistence.
  7. Keep the result readable: group automatic/safety skills separately from explicit skills when the inventory is long.

If the catalog is unavailable, enumerate installed skill folders containing SKILL.md, read their frontmatter, and mark only skills whose metadata explicitly says user activation is required. State that marker accuracy could not be verified against the catalog.

Marker Legend

  • * skill-name: requires direct user invocation or enabling.
  • skill-name: selected automatically by typed task conditions or mandatory safety routing.

The star describes activation, not quality, risk, or installation status.

Supported Commands

Always include commands that exist in the installed pack:

  • --help: show this inventory and command guide.
  • --quizme: toggle exhaustive conversation-local clarification.
  • --quizme --mc: prefer multiple-choice clarification.
  • --quizme --one-at-a-time: ask one adaptive question per round.
  • --quizme --confirm: require confirmation of the final task contract.
  • --quizme --record: record the approved contract when a suitable artifact is authorized; implies confirmation.
  • --internal-lang on|off: toggle private compact notation.
  • --internal-lang --response on|off: toggle compact user-facing notation separately.
  • --clean-slate: enable the fresh-context boundary.
  • --clean-slate off: disable the fresh-context boundary.
  • --ill-run-scripts: toggle the user-run-scripts pause; its state persists across conversations.

Do not advertise a command whose corresponding installed skill or host policy is absent.

Runtime Conformance

Use scripts/runtime_adapter.py as the executable reference for startup, exact-command parsing, and persisted mode transitions. It emits a message digest and a router-ready descriptor patch while atomically storing only explicit mode booleans in the host preference file. A host may implement an equivalent adapter, but must pass the conformance tests and must not claim support from prose-only behavior.

Output Contract

Return:

  1. marker legend
  2. complete active-skill inventory
  3. supported commands and short descriptions
  4. a note that skill activation never grants write, commit, push, publish, deploy, delete, message, or other external authority

Do not create or update any file merely because help was requested.

reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 467kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 319kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 319kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 319kUpdated Claude CodeCodexVerifiedInstall