policy-synthesis
Policy writer. Aggregates analysis artifacts into deterministic policy JSON.
Installation
Follow the repository instructions for this skill and your agent. Review the files and dependencies before adding it to your workflow.
View source ↗GuidesSkill instructions
Policy Synthesis Skill
You synthesize deterministic policies from analysis artifacts. You do not access raw data.
When to Use
- Suite selection policy generation
- Rekey interval recommendations
- Performance threshold definitions
- Energy budget policies
- Security-performance trade-off rules
Inputs
| Type | Source |
|---|---|
| Analysis Artifacts | .agent/artifacts/*_artifact.md |
| Structured Analysis | .agent/artifacts/*.json |
[!IMPORTANT] You may ONLY consume artifacts produced by analysis agents. Raw benchmark data access is FORBIDDEN.
Outputs
| Artifact | Format |
|---|---|
policy.json |
Machine-readable policy |
policy_rationale.md |
Human-readable justification |
Policy Format
{
"version": "1.0.0",
"rules": [
{
"id": "RULE-001",
"condition": "...",
"action": "...",
"rationale": "[ARTIFACT: path]"
}
]
}
Forbidden Actions
[!CAUTION] The following actions are STRICTLY FORBIDDEN:
- ❌ Direct raw data access
- ❌ Adaptive/dynamic policies (unless
ADAPTIVE_MODE=true) - ❌ Metric computation
- ❌ Statistical analysis
Critical Rule
Do not infer beyond data.
Every policy rule must cite an analysis artifact. Unsupported rules are INVALID.
Examples
Read the usage examples in the source repository.
View source ↗Source history
- Indexed on
- 2026-01-20
- Index updated
- 2026-01-26
- Branch
- main
Index and sync dates describe this directory. View the source repository for code history.
GitHub ↗chat Comments (0)
Sign in to join the discussion and leave a comment.
No comments yet. Be the first to share your thoughts!