Skip to content

farhanhidayatulhaq-max/GAMUKIS / .agents/skills/teamwork-dispatch

teamwork-dispatch

Verified

Use when delegating or accepting cross-agent tasks (Codex, Antigravity, Copilot, OpenClaw, etc.), writing DISPATCH files, establishing file boundaries and ownership, or verifying dispatch completeness before starting work.

Teamwork Dispatch

Delegating a Task (Lead)

  1. Copy .agents/templates/DISPATCH.md to .agents/teamwork/<task-name>/DISPATCH.md.
  2. Complete all sections. The three most commonly omitted sections with the highest downstream cost:
    • Allowed to modify and Owned by another agent: a single file must have only one owner at any given time.
    • Deliverables: must be inspectable (files, findings with path:line numbers, contract decisions, verification results). "Implement everything" is not an acceptable deliverable.
    • Verification: specify commands permitted to run and checks that must NOT be executed.
  3. Assign the recipient's role based on capability rather than agent brand. Refer to .agents/ORCHESTRATION.md.
  4. Add or update the task entry row in .agents/teamwork/INDEX.md.
  5. If direct communication channels are absent, the dispatch file is the sole delivery mechanism. Do not mark "accepted" until the recipient confirms receipt in progress.md.

Accepting a Task

  1. Read AGENTS.md, DISPATCH.md, followed by the latest handoff.md and progress.md in the task directory.
  2. Inspect the git status and diff within the assigned file boundary.
  3. If discrepancies or ambiguities exist (contracts, ownership, scope), record the questions in handoff.md and halt that specific branch. Do not broaden scope or unilaterally alter contracts.
  4. Record formal acceptance in progress.md signed as agent · model · date.

Small Tasks

Dispatches may be concise, but must still explicitly state the objective, deliverables, and ownership boundaries.

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: 468kUpdated Claude CodeCodexUnverifiedInstall
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