hoatv2211/GameStudio-CodexKIT / skills/studio-handoff
studio-handoff
UnverifiedUse when pausing, transferring, or reactivating game-studio work and a durable handoff must capture branch, goal, scope, files, commands, Verified Snapshot Unverified BLOCKED facts, failures, decisions, next actions, and a reactivation prompt.
Studio Handoff
Overview
Create a restartable state transfer that distinguishes verified work from snapshots, hypotheses, and blockers.
When to use
Use when a session ends, ownership changes, work becomes blocked, a subsystem crosses teams, or another agent must continue without repeating discovery.
When NOT to use
Do not create a handoff as a substitute for running available verification or resolving a small task that can be completed safely now.
Required inputs and context discovery
Collect repository/path, branch, goal, owned and excluded scope, current Git state, files touched, commands run, artifacts, decisions, failures, open hypotheses, dependencies, and next actions.
Safety and risk level
Writing a handoff is low-risk. Preserve existing project handoff conventions, never expose credentials, and never rewrite unrelated history or other sessions’ ownership.
Workflow
- Re-read the task packet, applicable instructions, and current Git status. Completion criterion: the handoff reflects current state rather than memory.
- List files touched and commands run with exact results. Completion criterion: every PASS claim maps to fresh evidence.
- Separate Verified, Snapshot, Unverified, and BLOCKED facts. Completion criterion: no hypothesis appears under Verified.
- Record failures, decisions, ownership boundaries, and do-not-touch paths. Completion criterion: the next session can avoid conflicting or unsafe work.
- Provide three to seven ordered next actions and a reactivation prompt. Completion criterion: another session can start with one bounded command or inspection.
Evidence and output contract
Produce HANDOFF.md with repository/path, branch, goal, scope ownership/do-not-touch, files touched, commands/tests, Verified results, Snapshot assumptions, Unverified hypotheses, BLOCKED items, failures, decisions, next actions, and Reactivation prompt.
Handoff contract
The output is the handoff contract. Keep exact paths, symbols, ports, error strings, and commands intact; sanitize secrets and raw credentials.
Pitfalls and anti-rationalization
- Do not say “tests pass” without the command and fresh output.
- Do not omit dirty or untracked files.
- Do not bury blockers in prose.
- Do not prescribe next actions that require unavailable permissions without labeling them BLOCKED.
Verification checklist
- Repository, branch, goal, and scope are current.
- Files and commands are exact.
- Verified, Snapshot, Unverified, and BLOCKED are separate.
- Failures and decisions are explicit.
- Three to seven next actions and a Reactivation prompt exist.
References and scripts
Follow the project's existing HANDOFF.md and .agents/CONTRACT.md when present; otherwise use the handoff contract in AGENTS.md.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| 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: 467k | Updated | Claude CodeCodex | Verified | Install |
| 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: 319k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 319k | Updated | Claude CodeCodex | Verified | Install |
| 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 available | Stars: 319k | Updated | Claude CodeCodex | Verified | Install |