list_rules

List every rule loaded for the current repository. Each entry includes id, name, severity, category, scope, source (builtin / repo / override), and a one-line description.

Use the id to fetch full markdown documentation via explain_rule.

Input schema

{ "type": "object", "properties": {} }

No arguments.

Output

A JSON array of rule summaries, wrapped in the MCP content envelope:

[
  {
    "id": "builtin.size.fn-length",
    "name": "Function length",
    "description": "Functions whose body spans more than the configured number of lines.",
    "severity": "warn",
    "category": "size",
    "scope": "file",
    "source": "builtin",
    "languages": ["rust", "python", "go", "java", "typescript", "tsx", "javascript"],
    "enabled": true,
    "tags": ["size", "complexity"]
  },
  {
    "id": "project.no-unwrap",
    "name": "No unwrap() in production code",
    "description": "Disallow .unwrap() outside test files.",
    "severity": "warn",
    "category": { "custom": "panics" },
    "scope": "file",
    "source": "repo",
    "languages": ["rust"],
    "enabled": true,
    "tags": ["rust", "panics"]
  }
]

The fields mirror rule frontmatter. The body field is intentionally omitted — it’s often long, and most callers just want metadata. Use explain_rule to fetch a body.

When to call

The agent is unsure which rules apply and wants to filter by category or language.
After authoring a new rule, to confirm it loaded.
As context for an explain_rule follow-up — the agent picks the id here, then asks for details.

Useful as a first call in a session — the agent learns what it’s being graded against before it edits anything.

Examples

{ "name": "list_rules", "arguments": {} }

Errors

MCP error code	Cause
`-32603` (internal error)	Failed to load `.sextant/rules/*/.md` (malformed YAML, IO error).