panjose/Co-Scientist / skills/run-policy-router
run-policy-router
VerifiedInterpret the user's high-level research intent into `RUN_POLICY.yaml` and `state/POLICY_DECISION.json`.
run-policy-router
Goal:
- Interpret the user's high-level research intent into
RUN_POLICY.yamlandstate/POLICY_DECISION.json.
Inputs:
- one run directory root
- run-local
input.md - optional
RUN_POLICY.yaml - research brief and operator constraints
Outputs:
- updated
RUN_POLICY.yaml - updated
state/POLICY_DECISION.json
Context Loading:
- Open
skills/shared-references/schema-index.md. - Read
packages/agent_contracts/policy.pyand confirm the exactRunPolicyContractplusPolicyDecisionContractshapes before writingRUN_POLICY.yamlorstate/POLICY_DECISION.json. - Read
input.mdplus any operator constraints that should influence the high-level routing axes.
Execution Contract:
- The router must stay at the high-level policy layer and must not invent raw numeric mechanics values directly.
- Allowed routing axes are documented in
../shared-references/policy-contract.md. - Downstream numeric settings are resolved separately into
state/RESOLVED_RUN_CONFIG.json. - This skill owns canonical policy artifact emission only:
RUN_POLICY.yamlstate/POLICY_DECISION.json
- This skill does not own numeric mechanics resolution, direct strategy routing, or downstream evolution-state writes.
Execution Steps:
- Open
skills/shared-references/schema-index.md, then readpackages/agent_contracts/policy.pybefore writingRUN_POLICY.yamlorstate/POLICY_DECISION.json. - Interpret the research brief into the allowed policy axes only.
- Write canonical
RUN_POLICY.yamlandstate/POLICY_DECISION.json. - Do not invent raw numeric mechanics values that belong in
state/RESOLVED_RUN_CONFIG.json. - Validate the emitted policy artifacts.
Completion Rule:
- This skill is complete only when
RUN_POLICY.yamlandstate/POLICY_DECISION.jsonhave been written in canonical form and remain consistent with the selected high-level policy axes.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |