Agent skill
localsetup-agentic-prd-batch
Agentic PRD batch: when user says 'process PRDs' or 'run batch from PRD folder'; implement per spec; update status; write outcome; reference PRD schema + external-agent guide. Use when editing .agent/queue/**, prds/**, *.prd.md.
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/localsetup-agentic-prd-batch
Metadata
Additional technical details for this skill
- version
- 1.3
SKILL.md
Agentic PRD batch
When the user says process PRDs, run batch from PRD folder, process the queue, or similar, or when editing files in .agent/queue/ with intent to implement:
- Locate specs: Look for PRD/spec files in the configured path (e.g.
.agent/queue/or structuredin/under queue root). Exclude README, INDEX, SPEC-TEMPLATE. Filter by front matterstatus == ready(orin-progressif resuming). Sort by priority (high first), then filename date (oldest first). Version mismatch: If a spec haslocalsetup_framework_versionand it differs from repo root VERSION, surface a warning per protocol; do not silently ignore. - Implement per spec: Load each spec; follow Implementation steps and Acceptance criteria; satisfy Verification plan and Rollback plan. Use PRD schema and _localsetup/docs/PRD_SCHEMA_EXTERNAL_AGENT_GUIDE.md for format and outcome template.
- Update status: Set spec
status: in-progresswhen starting; setstatus: done(orblocked) when finished. Update queue INDEX if present. - Write outcome: Append outcome block per spec (e.g.
## Outcomeor platform-specific section name, and/oraq_outcomeYAML) with branch, commit SHA, files changed, verification, rollback command. See PRD_SCHEMA_EXTERNAL_AGENT_GUIDE for template. - Clean-tree invariant: Before marking a spec done, ensure repo is clean (no modified tracked files except intended commits; untracked queue specs allowed). Commit or revert as needed.
- External confirmation: If spec front matter includes
external_confirmation: acknowledgedorimpact_review: confirmed_by: external_agent, agent may skip human impact confirmation; otherwise follow guardrails (impact summary + user confirmation for big/destructive changes).
Reference
- _localsetup/docs/AGENTIC_AGENT_Q_PATTERN.md - queue pattern (flat and structured layout).
- _localsetup/docs/PRD_SCHEMA_EXTERNAL_AGENT_GUIDE.md - spec format, front matter, outcome template, clarification protocol.
- _localsetup/docs/AGENTIC_AGENT_TO_AGENT_PROTOCOL.md - agent-to-agent handoff (ACTIVE); pre-ship gate before ship to peer.
- _localsetup/tools/agentq_transport_client/ - version stamp CLI:
python agentq_cli.py stamp-prd <path>.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?