메뉴
Pre-action boundary checking — validates agent tool calls against declared capabilities and task contracts
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | action-validator |
| description | Pre-action boundary checking — validates agent tool calls against declared capabilities and task contracts |
| scope | core |
| user-invocable | false |
Advisory pre-action validation layer that checks agent tool calls against declared capabilities, file access scope (R002), and task contracts before execution. Inspired by AutoHarness (Google DeepMind) — enforcing action-space legality at agent boundaries.
This skill does NOT block actions (R021 advisory-first model). It emits warnings when agents attempt operations outside their declared scope.
| Check | What | Against |
|---|---|---|
| Tool scope | Tool being called | Agent's tools frontmatter list |
| File scope | File path in Write/Edit | R002 file access rules |
| Domain scope | Target file extension | Agent's domain frontmatter |
| Task contract | Operation type | Task description constraints |
--- [Action Validator] Scope warning ---
Agent: {agent-name}
Tool: {tool-name}
Target: {file-path}
Issue: {description}
Declared scope: {agent's declared tools/domain}
💡 Suggestion: {recommended action}
---
| System | How |
|---|---|
| PreToolUse hooks | Optional hook to check tool calls (advisory only) |
| pipeline-guards | Complements pipeline stage gates |
| adversarial-review | Provides action-space-legality criterion |
| R002 (Permissions) | Validates against declared file access rules |
| R010 (Orchestrator) | Orchestrator validates subagent scope claims |
For high-repetition agents (e.g., mgr-gitnerd commit workflows), capture validated decision paths as reusable policies:
policy_cache:
agent: mgr-gitnerd
action: git-commit
validated_steps:
- tool: Bash
pattern: "git add *"
verdict: allow
- tool: Bash
pattern: "git commit *"
verdict: allow
- tool: Bash
pattern: "git push *"
verdict: warn_confirm
Policy caching reduces redundant LLM calls for well-understood workflows. Policies are advisory — the orchestrator may override.
When a synthesized harness exists for an agent (.codex/outputs/harnesses/{agent-name}-*.yaml), action-validator can use it for enhanced validation:
| Mode | Source | Behavior |
|---|---|---|
| Advisory (default) | Prompt-based checks | Emit warnings only |
| Code-verified | harness-synthesizer output | Run harness validation code, emit advisory results |
| Hard-enforce (opt-in) | harness-synthesizer --hard-enforce | Block invalid actions (requires explicit opt-in, see R021) |
To generate a harness for an agent: /harness-synthesizer --agent {name} --mode verifier
Code harness validation is additive — it supplements prompt-based checks, not replaces them.
This skill is an advisory layer, not a hard enforcement mechanism:
Auto-detect project context and optimize harness — deactivate unused agents/skills, suggest missing experts, generate project profile
Multi-LLM adversarial consensus loop — 3+ LLMs compete to find flaws in designs/specs until unanimous agreement is reached
Monitor Claude Code releases and auto-generate GitHub issues for each new version
Execute OpenAI Codex CLI prompts and return results
YAML-based DAG workflow engine with topological execution and failure strategies
Routes data engineering tasks to the correct DE expert agent. Use when user requests data pipeline design, DAG authoring, SQL modeling, stream processing, or warehouse optimization.