Menu
Final package QA checklist for clarity, safety, and import readiness
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Safely edit an existing Agent Hub asset with asset-type-specific checks and follow-up verification.
Create or revise the core assets for an Agent Hub agent, including soul, bundle, and optional profile.
Draft or update a portable Agent Hub skill with clean frontmatter and a practical workflow.
Use when implementation is complete on a normal feature branch and you need to verify, clean up commits, rebase, and either merge locally or create a PR without using a worktree-heavy workflow.
Use after implementing a feature, fixing a bug, or reaching a wave checkpoint when you need to actively test the behavior by running the app, exercising the flow, and confirming the change actually works.
Use when you have a written implementation plan to execute in a separate session with review checkpoints
| name | hr-final-check |
| description | Final package QA checklist for clarity, safety, and import readiness |
| audience | HR verifier subagent |
| license | MIT |
| compatibility | opencode >= 0.1 |
| metadata | {"domain":"agent-qa","version":"1.0"} |
Verify that a staged package is understandable and safe before the human operator approves it.
All of the following must be true:
All of the following must be true:
Write final-checklist.md and end with a human-facing status:
READY FOR HUMAN CONFIRMATIONREADY WITH CAVEATSNOT READYKeep final-checklist.md compact and explicit. Use this shape:
## Final Checklist - <package-id>
| Check | Result |
|---|---|
| import root is agenthub-home/ | pass/fail |
| no unsupported concepts | pass/fail |
| soul skill/tool sets declared | pass/fail |
| descriptions are operator-readable | pass/fail |
| MCP registrations resolve to staged servers or blocker | pass/fail |
| handoff clearly separates test/use/promote | pass/fail |
| model preferences were confirmed before assembly | pass/fail |
| final names were user-confirmed | pass/fail |
| specialized work was delegated | pass/fail |
| staged model ids confirmed via opencode environment | pass/fail |
| profile defaultAgent matches bundle agent.name | pass/fail |
| team includes at least one primary, non-hidden agent | pass/fail |
| default opencode agent policy confirmed | pass/fail |
| default-profile preference recorded if present | pass/fail |
| no host project mutations | pass/fail |
overall: READY FOR HUMAN CONFIRMATION | READY WITH CAVEATS | NOT READY
blocker: <description or none>
The package cannot be marked ready unless the verifier confirms:
agenthub hr <profile>, (c) promote command agenthub promote <package-id><package-root>/agenthub-home and is described as advanced/manual onlyskills/ directory or are explicitly rejected as missing blockerspython3 $HR_HOME/bin/validate_staged_package.py $HR_HOME/staging/<package-id> passespromotion_preferences.set_default_profile is present, it is consistentmcp/*.json files, the required mcp-servers/ implementation files, and mcp-servers/package.json when runtime dependencies are neededdefaultAgent, that value exactly matches one of the staged bundles' agent.name values (not just the bundle filename)team-only profile keeps at least one such primary agent available to the operatorIf bundle metadata contains fake runtime keys such as optional_skills or runtime_conditional_skills, mark the package NOT READY until they are removed or rewritten as plain documentation outside runtime bundle semantics.