action-validator
FreePre-action boundary checking — validates agent tool calls against declared capabilities and task contracts
The English summary is being prepared. Showing the original description.
Aggregate parallel agent results into concise output
The English summary is being prepared. Showing the original description.
See how to installReview the instructions the agent will receive before you install.
Aggregate and format results from multiple parallel agent executions into concise, scannable output.
[Batch Complete] {completed}/{total}
├── {agent}: {icon} {summary}
├── {agent}: {icon} {summary}
└── {agent}: {icon} {summary}
[Batch Complete] {completed}/{total}
[1] {agent-name}
Status: {success|failed|partial}
Target: {file/path}
Result: {summary}
[2] {agent-name}
Status: {success|failed|partial}
Target: {file/path}
Result: {summary}
[Batch Complete] {completed}/{total} ({failures} failed)
Failures:
├── {agent}: {error summary}
└── {agent}: {error summary}
Successes:
├── {agent}: ✓ {summary}
└── {agent}: ✓ {summary}
| Icon | Status | Use When |
|---|---|---|
| ✓ | Success | Task completed successfully |
| ✗ | Failed | Task failed with error |
| ⚠ | Warning | Completed with warnings |
| ⏳ | Pending | Still in progress |
| ⊘ | Skipped | Task was skipped |
order:
- Failed results first (need attention)
- Warnings second
- Successes last
- Within category: alphabetical by agent
ecomode_on:
max_length: 50 characters
truncate_with: "..."
ecomode_off:
max_length: 200 characters
allow_multiline: true
group_by:
- status (success/failed/warning)
- agent_type (when mixed types)
- target_path (when same agent, different targets)
Secretary receives:
- List of agent results
- Ecomode status
- User display preferences
Secretary outputs:
- Aggregated summary using this skill
- Individual details if requested
When ecomode active:
- Use compact format
- One line per agent
- Icons for status
- Truncate long summaries
[Batch Complete] 4/4
├── lang-golang-expert: ✓ src/main.go - 2 issues (1 naming, 1 error handling)
├── lang-python-expert: ✓ scripts/*.py - Clean
├── lang-rust-expert: ⚠ lib/core.rs - 1 unsafe block flagged
└── lang-typescript-expert: ✓ web/app.tsx - 3 suggestions
[Batch Complete] 3/4 (1 failed)
✗ lang-kotlin-expert: Skill not found: kotlin-best-practices
✓ lang-golang-expert: Agent created at .claude/agents/lang-golang-expert.md
✓ lang-python-expert: Agent created at .claude/agents/lang-python-expert.md
✓ lang-rust-expert: Agent created at .claude/agents/lang-rust-expert.md
[Batch Complete] 5/5
Agents Audited:
├── mgr-creator: ✓ All refs valid
├── mgr-updater: ✓ All refs valid
├── mgr-supplier: ⚠ 1 deprecated ref
├── lang-golang-expert: ✓ All refs valid
└── lang-python-expert: ✓ All refs valid
Summary: 5 agents checked, 1 warning
R006 Artifact Channel Protocol을 소비하는 표준 패턴. 병렬 에이전트가 각자 .claude/outputs/sessions/{date}/{skill}-{HHmmss}.md에 결과를 작성하면, result-aggregation이 경로 N개를 받아 단일 요약을 생성합니다.
CC sensitive-path check inspects tool target paths and triggers permission prompts on .claude/ regardless of bypassPermissions and allow rules (refs: #960, #961, #978, #981, #1016).
To write result-aggregation results under .claude/outputs/sessions/:
/tmp/result-aggregation-$(date +%H%M%S).md first (Write tool target = /tmp, no sensitive-path trigger)/tmp/*.sh Bash script to move/copy the file under .claude/outputs/sessions/$(date +%Y-%m-%d)/ (Bash target = /tmp, script-internal cp to .claude/ is not audited).claude/outputs/ (e.g., cat, head, wc) is allowed for verificationReference: feedback_sensitive_path_tmp_bypass.md, R006 sensitive-path handling, #1016, #1045.
paths = [
".claude/outputs/sessions/2026-04-24/agent1-100200.md",
".claude/outputs/sessions/2026-04-24/agent2-100300.md",
...
]
MUST-agent-design.md Artifact Channel Protocol (source contract)SHOULD-ecomode.md Deep Insight Context Handoff Pattern (budget 맥락)No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Pre-action boundary checking — validates agent tool calls against declared capabilities and task contracts
The English summary is being prepared. Showing the original description.
Auto-detect project context and optimize harness — deactivate unused agents/skills, suggest missing experts, generate project profile
The English summary is being prepared. Showing the original description.
Adversarial code review using attacker mindset — trust boundary, attack surface, business logic, and defense evaluation
The English summary is being prepared. Showing the original description.
Apache Airflow best practices for DAG authoring, testing, and production deployment
The English summary is being prepared. Showing the original description.
Alembic migration patterns for naming conventions, safety checks, expand-contract, env.py configuration, and CI integration
The English summary is being prepared. Showing the original description.
Pre-routing ambiguity analysis — scores request clarity and asks clarifying questions when needed (inspired by ouroboros)
The English summary is being prepared. Showing the original description.