cursor-doctor
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@cursor-doctoraudit the project and generate the prescription plan"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Cursor Doctor
A Node.js/TypeScript CLI that audits a project for documentation, rules, skills, and related knowledge artifacts. It compares what exists against a reference concept catalog and generates a prescription plan (cursor-doctor-prescription-plan.md) with frontmatter todos for missing or incomplete elements.

Install
Use directly via npx:
npx -y @ajgreyling/cursor-doctor@latest generateOr install globally:
npm install -g @ajgreyling/cursor-doctor
cursor-doctor generateRelated MCP server: ProofGate
Requirements
Node.js
>=18npm (or
npx)
CLI commands
Generate prescription plan (default)
cursor-doctor generate [--project <path>] [--out <filename>]If no command is provided, generate is used by default:
cursor-doctor --project /path/to/projectAudit only
cursor-doctor audit [--project <path>] [--json]Read a discovered document
cursor-doctor read --file <path> [--project <path>]Help
cursor-doctor helpFeatures
Audits discovered artifacts against a 23-concept catalog across six tiers.
Detects
found,incomplete, andmissingconcepts with reason details.Generates a markdown prescription plan with machine-readable YAML todos.
Supports read-only JSON audits for CI/reporting workflows.
Reads project files directly for quick artifact inspection (
readcommand).
README completeness checklist
project name or title
setup/install steps
features or capabilities
how to run or use
Operational workflow
Use this sequence for a full doctor cycle:
Run
cursor-doctor generate(ornpx -y @ajgreyling/cursor-doctor@latest generate).Open
cursor-doctor-prescription-plan.mdand work through frontmatter todos in order.Create/update missing artifacts from each todo
prompt.Complete the final
artifact-synctodo by updating affected docs/rules/skills.Re-run
cursor-doctor generateto confirm gaps are resolved.
Doctor command in Cursor
If you want a doctor / cursor doctor skill in Cursor, point that skill to run the CLI (npx -y @ajgreyling/cursor-doctor@latest generate) and then work from the frontmatter todos in the generated plan.
Tracked Cursor artifacts
These files are committed in this repo so the doctor workflow stays consistent:
File | Purpose |
Skill for "doctor" / "cursor doctor" that runs the CLI and works from plan todos. | |
Rule applied when editing the prescription plan or rules/skills; complete the final Update Artifacts step. | |
Rule (always apply): the cursor-doctor skill must exist at |
Concept catalog
The tool checks for 23 concepts in 6 tiers. Artifacts are matched by file patterns (for example README.md, .cursor/rules/*.mdc) and by content signals so differently named files can still satisfy a concept.
Tier 1 – Foundation: README, project-overview rule, file-locations rule, agent-workflow rule.
Tier 2 – Architecture: Architecture doc, data model doc, API spec (for example OpenAPI).
Tier 3 – Development conventions: Backend, frontend, database, and lessons-learnt rules.
Tier 4 – AI knowledge and skills: AGENTS doc, knowledge-base skill, code-generation skill, operational skills, dependency-management skill.
Tier 5 – Quality and documentation: Testing spec, accessibility doc, style guide, UI documentation.
Tier 6 – Infrastructure: MCP config, container config, CI/CD config.
Final – Artifact sync: “Update Artifacts” is always the final todo.
Project-specific concept profiles
cursor-doctor supports per-project concept customization with this precedence:
package.jsonfield:cursorDoctorfallback file:
.cursor/cursor-doctor.concepts.yamlbuilt-in default catalog
This makes it possible to enforce domain-specific docs (for example, custom architecture/data-workflow docs) without changing the CLI code.
package.json profile example
{
"cursorDoctor": {
"includeDefaultCatalog": true,
"concepts": [
{
"id": "domain-workflows-doc",
"name": "Domain Workflows",
"tier": 2,
"intent": "Domain workflow coverage and execution paths",
"filePatterns": ["docs/WORKFLOWS.md"],
"contentSignals": ["workflow", "flowchart", "sequence"],
"completenessChecks": ["mermaid", "actors", "happy path", "failure path"],
"generationPrompt": "Create docs/WORKFLOWS.md with mermaid diagrams and workflow coverage."
}
]
}
}YAML fallback example (non-Node repos)
Create .cursor/cursor-doctor.concepts.yaml:
includeDefaultCatalog: true
concepts:
- id: domain-state-machines-doc
name: Domain State Machines
tier: 2
intent: Lifecycle status transitions and guards
filePatterns:
- docs/STATE_MACHINES.md
contentSignals:
- state machine
- transition
- status
completenessChecks:
- mermaid
- states
- transitions
generationPrompt: Create docs/STATE_MACHINES.md with mermaid state diagrams.Override behavior
Concept IDs are unique keys.
A project profile concept with an existing ID overrides the built-in concept.
A new ID appends a new concept.
The final
artifact-synctodo is always enforced.
Build and local run
npm install
npm run build
node dist/index.js generateAudit-only examples
# Human-readable summary
node dist/index.js audit
# JSON output for automation
node dist/index.js audit --jsonPublishing
npm login # once, if needed
./publish.sh # build + npm publish
./publish.sh --dry-run # test without publishingDocumentation index
Core docs
Project docs (docs/)
Cursor skills (.cursor/skills/)
License
MIT
This server cannot be deployed
Maintenance
Related MCP Connectors
MCP Spec Compliance MCP — audits any MCP server.json against the official Model Context Protocol
Evidence-readiness MCP server: validate, audit, and score briefs, memos, and evidence packs.
MCP server for generating rough-draft project plans from natural-language prompts.
Scan any MCP server for tool-poisoning, security, auth & license. Trust score before install.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceMCP server that enables coding agents to check documentation claims against source code, detecting drift and suggesting fixes.14 npm2MIT
- AlicenseNot gradedqualityCmaintenanceEvidence-first delivery audit MCP server that evaluates task requirements against delivery evidence and returns a reproducible pass/needs_review/fail decision with a deterministic receipt.MIT
- AlicenseAqualityAmaintenanceMCP server that provides audit and safety-check tools for enterprise SDLC code integrity, enabling AI agents to scan workspaces for lifecycle gaps, mock-theater tests, DRY violations, and language-specific issues in shell, JavaScript/HTML, and Python.413 npmMIT
- AlicenseNot gradedqualityAmaintenanceMCP server that enables AI agents to search, fetch, and analyze a self-maintaining markdown knowledge base with provenance, drift detection, and canonical definitions.1MIT