How do I install the Co Scientist Doctor skill?

Run npx skillmds@latest add panjose/co-scientist-doctor-2 in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

What does the Co Scientist Doctor skill do?

co-scientist-doctor It is listed under Coding & Dev Tools on SkillMD.

Is Co Scientist Doctor safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Co Scientist Doctor?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Co Scientist Doctor free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Co Scientist Doctor?

panjose (@panjose) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Co Scientist Doctor by panjose — Agent Skill

Name: Co Scientist Doctor
Author: panjose

File contents

co-scientist-doctor

Goal:

Run the Co-Scientist environment diagnostics and explain any failing or warning checks to the user.

Expected input:

no arguments for the default project-local doctor

Execution steps:

Run:

python -m tools.host.claude_project_cli doctor

or, when machine-readable output is needed:

python -m tools.host.claude_project_cli doctor --format json

Read the emitted diagnostic output and summarize:
- Python version health
- Python dependency health
- active environment / conda or virtualenv status
- dashboard tooling (node, pnpm)
- dashboard install/build readiness
- project-local skill installation state
- how budget, iteration-policy, iteration-band, stop-policy, and human-checkpoint divide responsibility
- how dashboard ready URLs are resolved after a background bootstrap
- what inspect_state or validation blocked means operationally

If the dashboard checks warn, recommend:

pnpm --dir apps/dashboard install
pnpm --dir apps/dashboard build

If the dashboard checks pass, remind the user:
- fresh start, run, and resume commands try a background dashboard bootstrap
- python -m tools.host.claude_project_cli dashboard <run-dir> resolves the ready dashboard URL
- runs/<run_id>/dashboard/LINKS.md is the human-readable dashboard receipt
- budget controls per-round intensity, while iteration-policy controls semantic vs capped stopping
- human-checkpoint=auto means completion-driven runs should not pause after every evolution round
- inspect_state means review/ranking/evolution artifacts disagree and need repair before resume or overview

If the project-local skills are not installed, recommend:

powershell -File tools/install/install_co_scientist.ps1

Rules:

Prefer conda run -n <env> python ... over conda activate when the user is running commands through Claude Code shells.
Treat doctor warnings as actionable guidance, not as fatal runtime errors unless the check status is fail.

---name: co-scientist-doctor-2description: co-scientist-doctor---# co-scientist-doctorGoal:- Run the Co-Scientist environment diagnostics and explain any failing or warning checks to the user.Expected input:- no arguments for the default project-local doctorExecution steps:1. Run: ```bash python -m tools.host.claude_project_cli doctor ``` or, when machine-readable output is needed: ```bash python -m tools.host.claude_project_cli doctor --format json ```2. Read the emitted diagnostic output and summarize: - Python version health - Python dependency health - active environment / conda or virtualenv status - dashboard tooling (`node`, `pnpm`) - dashboard install/build readiness - project-local skill installation state - how `budget`, `iteration-policy`, `iteration-band`, `stop-policy`, and `human-checkpoint` divide responsibility - how dashboard ready URLs are resolved after a background bootstrap - what `inspect_state` or `validation blocked` means operationally3. If the dashboard checks warn, recommend: ```bash pnpm --dir apps/dashboard install pnpm --dir apps/dashboard build ```4. If the dashboard checks pass, remind the user: - fresh `start`, `run`, and `resume` commands try a background dashboard bootstrap - `python -m tools.host.claude_project_cli dashboard <run-dir>` resolves the ready dashboard URL - `runs/<run_id>/dashboard/LINKS.md` is the human-readable dashboard receipt - `budget` controls per-round intensity, while `iteration-policy` controls semantic vs capped stopping - `human-checkpoint=auto` means completion-driven runs should not pause after every evolution round - `inspect_state` means review/ranking/evolution artifacts disagree and need repair before resume or overview5. If the project-local skills are not installed, recommend: ```powershell powershell -File tools/install/install_co_scientist.ps1 ```Rules:- Prefer `conda run -n <env> python ...` over `conda activate` when the user is running commands through Claude Code shells.- Treat doctor warnings as actionable guidance, not as fatal runtime errors unless the check status is `fail`.

Frequently asked questions

Co Scientist Doctor