co-scientist-resume

co-scientist-resume

Goal:

• Resume one interrupted Co-Scientist run from Claude Code.

Expected input:

• one run directory such as runs/test1

Execution steps:

1. Resolve the run directory and confirm that input.md and state/ artifacts exist.

2. Run:

   python -m tools.host.claude_project_cli resume <run-dir> --skill co-scientist-pipeline
   

3. Read the persisted state artifacts:

   • runs/<run_id>/state/PIPELINE_STATE.json
   • runs/<run_id>/state/CURRENT_STAGE.json
   • runs/<run_id>/state/HOST_AGENT_HANDOFF.json

4. Read the CLI JSON result and the run-local dashboard receipt artifacts:

   • runs/<run_id>/dashboard/LINKS.md
   • runs/<run_id>/dashboard/LINKS.json

5. If the CLI JSON contains dashboardLinks:

   • If dashboard.status is running, return dashboardLinks.dashboard as the primary dashboard URL and include the deep links.
   • If dashboard.status is starting, immediately run:

   python -m tools.host.claude_project_cli dashboard <run-dir>
   

   • Read the refreshed CLI JSON result plus runs/<run_id>/dashboard/LINKS.md.
   • If runtime.status is now running, return the refreshed links.dashboard URL as the primary dashboard URL and include the deep links.
   • If runtime.status is still starting, tell the user that the dashboard is still booting, point them to runs/<run_id>/dashboard/LINKS.md, and include the retry command:

   /co-scientist-dashboard <run-dir>
   

6. Run the resume validator before continuing:

   python -m tools.validation.contract_validation <run-dir> --resume --skill co-scientist-pipeline
   

7. Resume only the incomplete phases. Do not restart completed phases when their artifacts remain valid.

8. Continue from the canonical skills/co-scientist-pipeline/SKILL.md flow. If the persisted route returns run_configuration, resume or rerun research-config before any generation work.

Rules:

• Resume decisions must be artifact-driven.
• Preserve existing manifest history and dashboard links.
• Treat validator failures as blocking until they are understood and fixed.
• If research_plan/RESEARCH_PLAN.json is missing or invalid, expect resume routing to return run_configuration instead of skipping directly to generation or evolution.
• Obey the persisted route exactly during resume: run_review resumes only review, run_insights resumes only insights, run_proximity resumes only proximity, run_ranking resumes only ranking, and continue_evolution may create at most one child before closing that child through review, proximity, ranking with ranking update receipt coverage, convergence, and one appended round receipt.
• Do not synthesize placeholder hypotheses, reviews, tournaments, proximity receipts, embeddings, or evolution-round receipts to make progress.
• If the required sub-skill or canonical tool cannot be executed, stop and report a resumable blocked state instead of writing low-information artifacts.
• A completed evolution round must be replayable from exactly one router decision, one evolved child, one review bundle, one proximity receipt, completed ranking artifacts with ranking update receipt coverage, one convergence update, and one appended round receipt.
• Round receipt tournament refs must be child-owned and duplicate-free; do not copy later opponent-side lifetime refs from HYPOTHESIS.json into an earlier EVOLUTION_ROUNDS.jsonl record.
• If the effective run policy is iteration_policy = completion_driven and human_checkpoint = auto, do not ask for per-round confirmation during resume. Continue until a true terminal route, configured checkpoint boundary, or blocking state is reached.
• If you must stop before convergence or a terminal route, tell the user the run is paused, current convergence has not been reached, persisted state is resumable, and the next recommended action is continue evolution through resume or an explicit continue request.
• Treat runs/<run_id>/dashboard/LINKS.md as the human-readable dashboard receipt and runs/<run_id>/dashboard/LINKS.json as the machine-readable receipt.