feat: add docs drift check

docs/api-reference.md

@@ -1,128 +0,0 @@
-# API Reference
-
-## Core Endpoints
-
-- `GET /`: server metadata.
-- `GET /health`: health probe.
-- `GET /metrics`: Prometheus metrics (when enabled).
-
-## OAuth Discovery and Token Exchange
-
-- `GET /.well-known/oauth-protected-resource`
-  - Returns OAuth protected resource metadata used by MCP clients.
-- `GET /.well-known/oauth-authorization-server`
-  - Returns OAuth authorization server metadata.
-- `POST /register`
-  - Registers an OAuth client and persists the client metadata.
-- `POST /oauth/token`
-  - Proxies OAuth authorization-code token exchange to Gitea.
-
-## MCP Endpoints
-
-- `GET /mcp/tools`: list tool definitions.
-- `GET /mcp` and `POST /mcp`: streamable HTTP transport.
-- `GET /mcp/sse` and `POST /mcp/sse`: MCP SSE transport alias.
-- `POST /mcp/tool/call`: direct tool-call endpoint.
-
-Authentication requirements:
-
-- MCP tool execution requires `Authorization: Bearer <token>`.
-- Missing or invalid tokens return `401` with:
-  - `WWW-Authenticate: Bearer resource_metadata="<absolute metadata url>", scope="read:repository"`
-
-Scope requirements:
-
-- Read tools require `read:repository`.
-- Write tools require `write:repository`.
-- Insufficient scope returns `403`.
-
-## Automation Endpoints
-
-- `POST /automation/webhook`: ingest policy-controlled webhook events.
-- `POST /automation/jobs/run`: run policy-controlled automation jobs.
-
-## Read Tools
-
-- `list_repositories`
-- `get_repository_info` (`owner`, `repo`)
-- `get_file_tree` (`owner`, `repo`, optional `ref`, `recursive`)
-- `get_file_contents` (`owner`, `repo`, `filepath`, optional `ref`)
-- `search_code` (`owner`, `repo`, `query`, optional `ref`, `page`, `limit`)
-- `list_commits` (`owner`, `repo`, optional `ref`, `page`, `limit`)
-- `get_commit_diff` (`owner`, `repo`, `sha`)
-- `compare_refs` (`owner`, `repo`, `base`, `head`)
-- `list_issues` (`owner`, `repo`, optional `state`, `page`, `limit`, `labels`)
-- `get_issue` (`owner`, `repo`, `issue_number`)
-- `list_pull_requests` (`owner`, `repo`, optional `state`, `page`, `limit`)
-- `get_pull_request` (`owner`, `repo`, `pull_number`)
-- `list_labels` (`owner`, `repo`, optional `page`, `limit`)
-- `list_tags` (`owner`, `repo`, optional `page`, `limit`)
-- `list_releases` (`owner`, `repo`, optional `page`, `limit`)
-- `list_pull_request_files` (`owner`, `repo`, `pull_number`, optional `page`, `limit`)
-- `list_pull_request_commits` (`owner`, `repo`, `pull_number`, optional `page`, `limit`)
-- `list_issue_comments` (`owner`, `repo`, `issue_number`, optional `page`, `limit`)
-- `list_branches` (`owner`, `repo`, optional `page`, `limit`)
-- `get_branch` (`owner`, `repo`, `branch`)
-- `get_release` (`owner`, `repo`, `release_id`)
-- `get_latest_release` (`owner`, `repo`)
-- `list_milestones` (`owner`, `repo`, optional `state`, `page`, `limit`)
-- `get_commit_status` (`owner`, `repo`, `sha`)
-- `list_org_repositories` (`org`, optional `page`, `limit`)
-- `list_organizations` (optional `page`, `limit`)
-- `get_repo_languages` (`owner`, `repo`)
-- `list_repo_topics` (`owner`, `repo`)
-
-## Write Tools (Write Mode Required)
-
-- `create_issue` (`owner`, `repo`, `title`, optional `body`, `labels`, `assignees`, `milestone`)
-- `update_issue` (`owner`, `repo`, `issue_number`, one or more of `title`, `body`, `state`, `milestone`)
-- `create_issue_comment` (`owner`, `repo`, `issue_number`, `body`)
-- `create_pr_comment` (`owner`, `repo`, `pull_number`, `body`)
-- `add_labels` (`owner`, `repo`, `issue_number`, `labels` by name)
-- `remove_labels` (`owner`, `repo`, `issue_number`, `labels` by name)
-- `assign_issue` (`owner`, `repo`, `issue_number`, `assignees`)
-- `create_label` (`owner`, `repo`, `name`, `color` hex e.g. `#00aabb`, optional `description`, `exclusive`)
-- `update_label` (`owner`, `repo`, `name`, one or more of `new_name`, `color`, `description`)
-- `create_pull_request` (`owner`, `repo`, `title`, `head`, `base`, optional `body`)
-- `create_release` (`owner`, `repo`, `tag_name`, optional `name`, `body`, `draft`, `prerelease`, `target`)
-- `edit_release` (`owner`, `repo`, `release_id`, one or more of `name`, `body`, `draft`, `prerelease`)
-- `create_branch` (`owner`, `repo`, `new_branch_name`, optional `old_branch_name`)
-- `create_milestone` (`owner`, `repo`, `title`, optional `description`, `due_on`)
-- `edit_issue_comment` (`owner`, `repo`, `comment_id`, `body`)
-
-Not supported by the dedicated tools by design: merge, branch/label/release deletion,
-force push, repo/admin management. Endpoints not covered above are reachable through the
-generic `gitea_request` escape hatch (subject to policy, write-mode, and a sensitive-path
-denylist) — see [Raw API Dispatch](raw-api.md).
-
-## Raw API Dispatch
-
-- `gitea_request` (`method`, `path`, optional `query`, `body`)
-  - Calls an arbitrary Gitea REST endpoint. `GET`/`HEAD` are reads; other methods are
-    writes and require write-mode plus a whitelisted repository. Admin/credential
-    endpoints are blocked unless `RAW_API_ALLOW_SENSITIVE=true`. See
-    [Raw API Dispatch](raw-api.md) for the two-layer policy model and full details.
-
-Note: `create_issue`, `add_labels`, and `remove_labels` accept label **names**; the
-server resolves them to Gitea label ids and returns a clear error for unknown labels.
-
-Note: the `milestone` argument on `create_issue`/`update_issue` accepts either a numeric
-milestone **id** or a milestone **title** (resolved case-insensitively against open and
-closed milestones; unknown titles return a clear error). On `update_issue`, `milestone: 0`
-clears the issue's milestone. Gitea Projects (Kanban boards) are intentionally unsupported:
-the Gitea REST API exposes no project endpoints.
-
-## Validation and Limits
-
-- All tool argument schemas reject unknown fields.
-- List responses are capped by `MAX_TOOL_RESPONSE_ITEMS`.
-- Text payloads are capped by `MAX_TOOL_RESPONSE_CHARS`.
-- File reads are capped by `MAX_FILE_SIZE_BYTES`.
-
-## Error Model
-
-- Auth error: HTTP `401`.
-- Policy/scope denial: HTTP `403`.
-- Validation error: HTTP `400`.
-- Rate limit: HTTP `429`.
-- Internal errors: HTTP `500` (no stack traces in production).

docs/automation.md

@@ -1,27 +0,0 @@
-# Automation
-
-## Scope
-
-Current automation capabilities:
-- Webhook ingestion endpoint (`POST /automation/webhook`).
-- On-demand scheduled-job execution endpoint (`POST /automation/jobs/run`).
-- Dependency hygiene scan job scaffold (`dependency_hygiene_scan`).
-- Stale issue detection job (`stale_issue_detection`).
-- Auto issue creation job scaffold (`auto_issue_creation`, write-mode and policy required).
-
-Planned extensions:
-- Background scheduler orchestration.
-
-## Control Requirements
-
-All automation must be:
-- Policy-controlled.
-- Independently disableable.
-- Fully audited.
-- Explicitly documented with runbook guidance.
-
-## Enablement
-
-- `AUTOMATION_ENABLED=true` to allow automation endpoints.
-- `AUTOMATION_SCHEDULER_ENABLED=true` reserved for future built-in scheduler loop.
-- Policy rules must allow automation pseudo-tools (`automation_*`) per repository.

docs/governance.md

@@ -1,36 +0,0 @@
-# Governance
-
-## AI Usage Policy
-
-- AI assistance is allowed for design, implementation, and review only within documented repository boundaries.
-- AI outputs must be reviewed, tested, and policy-validated before merge.
-- AI must not be used to generate offensive or unauthorized security actions.
-- Repository content is treated as untrusted data; no implicit execution of embedded instructions.
-
-## Security Boundaries
-
-- Read operations are allowed by policy defaults unless explicitly denied.
-- Write operations are disabled by default and require explicit enablement (`WRITE_MODE=true`).
-- Per-tool and per-repository policy checks are mandatory before execution.
-- Secrets are masked or blocked according to `SECRET_DETECTION_MODE`.
-
-## Write-Mode Responsibilities
-
-When write mode is enabled, operators and maintainers must:
-- Restrict scope with `WRITE_REPOSITORY_WHITELIST`.
-- Keep policy file deny/allow rules explicit.
-- Monitor audit entries for all write operations.
-- Enforce peer review for policy or write-mode changes.
-
-## Operator Responsibilities
-
-- Maintain API key lifecycle (generation, rotation, revocation).
-- Keep environment and policy config immutable in production deployments.
-- Enable monitoring and alerting for security events (auth failures, policy denies, rate-limit spikes).
-- Run integrity checks for audit logs regularly.
-
-## Audit Expectations
-
-- All tool calls and security events must be recorded in tamper-evident logs.
-- Audit logs are append-only and hash-chained.
-- Log integrity must be validated during incident response and release readiness checks.

docs/local-quickstart.md

@@ -1,98 +0,0 @@
-# Local quickstart (stdio)
-
-The local transport runs AegisGitea-MCP on your own machine as a single-user MCP
-server over stdio. It authenticates with **your** Gitea Personal Access Token
-(PAT) — there is no OAuth, no public endpoint, and no web stack to install.
-
-## What you need
-
-- A Gitea instance URL (`GITEA_URL`).
-- A Gitea Personal Access Token (`GITEA_TOKEN`) with least privilege:
-  - `read:repository`
-  - `write:repository` only if you intend to enable `WRITE_MODE`.
-- [`uv`](https://docs.astral.sh/uv/) (for `uvx`) or `pip`.
-
-## Run it
-
-With `uvx` (no install):
-
-```bash
-GITEA_URL=https://git.hiddenden.cafe \
-GITEA_TOKEN=<pat> \
-uvx aegis-gitea-mcp
-```
-
-With pip:
-
-```bash
-pip install aegis-gitea-mcp
-GITEA_URL=https://git.hiddenden.cafe GITEA_TOKEN=<pat> aegis-gitea-mcp
-```
-
-A local `.env` file is also supported — drop `GITEA_URL` and `GITEA_TOKEN` in it
-and just run `aegis-gitea-mcp`.
-
-If a required variable is missing the server exits with a clear message instead
-of a traceback.
-
-## Wire it into a client
-
-Claude Code:
-
-```bash
-claude mcp add aegis-gitea \
-  -e GITEA_URL=https://git.hiddenden.cafe \
-  -e GITEA_TOKEN=<pat> \
-  -- uvx aegis-gitea-mcp
-```
-
-Claude Desktop (`claude_desktop_config.json`):
-
-```json
-{
-  "mcpServers": {
-    "aegis-gitea": {
-      "command": "uvx",
-      "args": ["aegis-gitea-mcp"],
-      "env": {
-        "GITEA_URL": "https://git.hiddenden.cafe",
-        "GITEA_TOKEN": "<pat>"
-      }
-    }
-  }
-}
-```
-
-## What still applies locally
-
-The local adapter is single-user and trusts the PAT owner, so it skips the
-per-user repository-permission probe used by the public server. Everything else
-is identical to the server:
-
-- **Policy engine** (`policy.yaml`) — same allow/deny rules.
-- **`WRITE_MODE`** — off by default; writes are denied unless explicitly enabled
-  and whitelisted.
-- **`gitea_request`** full-API escape hatch — same write classifier, known-path
-  gate, and admin/credential denylist.
-- **Secret sanitization** of tool output.
-- **Tamper-evident audit log** — written to a per-user path when the container
-  default is not writable:
-  - Windows: `%LOCALAPPDATA%\aegis-gitea-mcp\audit.log`
-  - Linux/macOS: `$XDG_STATE_HOME/aegis-gitea-mcp/audit.log` or
-    `~/.local/state/aegis-gitea-mcp/audit.log`
-  - Override with `AUDIT_LOG_PATH`.
-
-## Enabling writes locally
-
-Writes are opt-in, exactly as on the server:
-
-```bash
-GITEA_URL=https://git.hiddenden.cafe \
-GITEA_TOKEN=<pat-with-write-repository> \
-WRITE_MODE=true \
-WRITE_REPOSITORY_WHITELIST=acme/app,acme/docs \
-uvx aegis-gitea-mcp
-```
-
-See [configuration.md](configuration.md) for the full variable reference and
-[write-mode.md](write-mode.md) for the write-mode model.

docs/observability.md

@@ -1,48 +0,0 @@
-# Observability
-
-## Logging
-
-- Structured JSON logs.
-- Request correlation via `X-Request-ID`.
-- Security events and policy denials are audit logged.
-
-### Structured event helpers
-
-`logging_utils` exposes reusable helpers so endpoints emit consistent,
-secret-safe structured events instead of ad-hoc inline logging:
-
-- `log_event(logger, level, event, **context)` — emit a named event with a
-  `context` mapping; keys in `SENSITIVE_CONTEXT_KEYS` (e.g. `token`,
-  `authorization`, `password`) are masked as `***`.
-- `log_nullable_field(logger, event, field, value)` — record whether a parsed
-  response field is `None` and its runtime type, without dumping its contents.
-- `sanitize_context(context)` — the masking primitive used by the above.
-
-The `context` mapping is serialized into the JSON log payload under a `context`
-key. These run at `DEBUG`, so they are silent unless `LOG_LEVEL=DEBUG`.
-
-`get_issue` is instrumented with these helpers (`get_issue.start`,
-`get_issue.payload_shape`, `get_issue.field_check`) to make nullable-field
-parsing failures diagnosable. The same pattern can be reused for other
-parsing-heavy endpoints (`get_pull_request`, `list_issues`, `get_commit_diff`).
-
-## Metrics
-
-Prometheus-compatible endpoint: `GET /metrics`.
-
-Current metrics:
-- `aegis_http_requests_total{method,path,status}`
-- `aegis_tool_calls_total{tool,status}`
-- `aegis_tool_duration_seconds_sum{tool}`
-- `aegis_tool_duration_seconds_count{tool}`
-
-## Tracing and Correlation
-
-- Request IDs propagate in response header (`X-Request-ID`).
-- Tool-level correlation IDs included in MCP responses.
-
-## Operational Guidance
-
-- Alert on spikes in 401/403/429 rates.
-- Alert on repeated `access_denied` and auth-rate-limit events.
-- Track tool latency trends for incident triage.

docs/policy.md

@@ -1,52 +0,0 @@
-# Policy Engine
-
-## Overview
-
-Aegis uses a YAML policy engine to authorize tool execution before any Gitea API call is made.
-
-## Behavior Summary
-
-- Global tool allow/deny supported.
-- Per-repository tool allow/deny supported.
-- Optional repository path allow/deny supported.
-- Write operations are denied by default.
-- Write operations also require `WRITE_MODE=true` and either:
-  - `WRITE_REPOSITORY_WHITELIST` match, or
-  - `WRITE_ALLOW_ALL_TOKEN_REPOS=true`.
-
-## Example Configuration
-
-```yaml
-defaults:
-  read: allow
-  write: deny
-
-tools:
-  deny:
-    - search_code
-
-repositories:
-  acme/service-a:
-    tools:
-      allow:
-        - get_file_contents
-        - list_commits
-    paths:
-      allow:
-        - src/*
-      deny:
-        - src/secrets/*
-```
-
-## Failure Behavior
-
-- Invalid YAML or invalid schema: startup failure (fail closed).
-- Denied tool call: HTTP `403` + audit `access_denied` entry.
-- Path traversal attempt in path-scoped tools: denied by validation/policy checks.
-
-## Operational Guidance
-
-- Keep policy files version-controlled and code-reviewed.
-- Prefer explicit deny entries for sensitive tools.
-- Use repository-specific allow lists for high-risk environments.
-- Test policy updates in staging before production rollout.

docs/todo.md

@@ -1,105 +0,0 @@
-# TODO
-
-## Phase 0 Governance
-
-- [x] Add `CODE_OF_CONDUCT.md`.
-- [x] Add governance policy documentation.
-- [x] Upgrade `AGENTS.md` as authoritative AI contract.
-
-## Phase 1 Architecture
-
-- [x] Publish roadmap and threat/security model updates.
-- [x] Publish phased TODO tracker.
-
-## Phase 2 Expanded Read Tools
-
-- [x] Implement `search_code`.
-- [x] Implement `list_commits`.
-- [x] Implement `get_commit_diff`.
-- [x] Implement `compare_refs`.
-- [x] Implement `list_issues`.
-- [x] Implement `get_issue`.
-- [x] Implement `list_pull_requests`.
-- [x] Implement `get_pull_request`.
-- [x] Implement `list_labels`.
-- [x] Implement `list_tags`.
-- [x] Implement `list_releases`.
-- [x] Add input validation and response bounds.
-- [x] Add unit/failure-mode tests.
-
-## Phase 3 Policy Engine
-
-- [x] Implement YAML policy loader and validator.
-- [x] Implement per-tool and per-repo allow/deny.
-- [x] Implement optional path restrictions.
-- [x] Enforce default write deny.
-- [x] Add policy unit tests.
-
-## Phase 4 Write Mode
-
-- [x] Implement write tools (`create_issue`, `update_issue`, comments, labels, assignment).
-- [x] Keep write mode disabled by default.
-- [x] Enforce repository whitelist.
-- [x] Ensure no merge/deletion/force-push capabilities.
-- [x] Add write denial tests.
-
-## Phase 5 Hardening
-
-- [x] Add secret detection + mask/block controls.
-- [x] Add prompt-injection defensive model (data-only handling).
-- [x] Add tamper-evident audit chaining and validation.
-- [x] Add per-IP and per-token rate limiting.
-
-## Phase 6 Automation
-
-- [x] Implement webhook ingestion pipeline.
-- [x] Implement on-demand scheduled jobs runner endpoint.
-- [x] Implement auto issue creation job scaffold from findings.
-- [x] Implement dependency hygiene scan orchestration scaffold.
-- [x] Implement stale issue detection automation.
-- [x] Add automation endpoint tests.
-
-## Phase 7 Deployment
-
-- [x] Harden Docker runtime defaults.
-- [x] Separate dev/prod compose profiles.
-- [x] Preserve non-root runtime and health checks.
-
-## Phase 8 Observability
-
-- [x] Add Prometheus metrics endpoint.
-- [x] Add structured JSON logging.
-- [x] Add request ID correlation.
-- [x] Add tool timing metrics.
-
-## Phase 9 Testing and Release Readiness
-
-- [x] Extend unit tests.
-- [x] Add policy tests.
-- [x] Add secret detection tests.
-- [x] Add write-mode denial tests.
-- [x] Add audit integrity tests.
-- [ ] Add integration-tagged tests against live Gitea (optional CI stage).
-- [ ] Final security review sign-off.
-- [ ] Release checklist execution.
-
-## Phase 10 Local Package & Safe Full Coverage (0.2.0)
-
-- [x] Extract transport-agnostic core + shared tool registry.
-- [x] Lock the core/web boundary with a no-fastapi import test.
-- [x] Add local stdio adapter (`stdio_app.py`) over the `mcp` SDK.
-- [x] Restructure packaging: core install + `[server]` extra + console scripts.
-- [x] Resource-type-aware authorization (repo/org/user/admin/misc), fail-closed.
-- [x] Classified `gitea_request`: write classifier + known-path gate + denylist.
-- [x] Authz matrix, write-mode bypass, classifier, and stdio adapter tests.
-- [x] `.gitea/workflows/publish.yml` (uv build + publish to Gitea registry on tag).
-- [ ] Make `list_organizations` user-scoped in service-PAT mode (`/users/{login}/orgs`)
-      so it can be allowed instead of denied. (TODO(authz))
-
-## Release Checklist
-
-- [ ] `make lint`
-- [ ] `make test`
-- [ ] Documentation review complete
-- [ ] Policy file reviewed for production scope
-- [ ] Write mode remains disabled unless explicitly approved

docs/write-mode.md

@@ -1,55 +0,0 @@
-# Write Mode
-
-## Threat Model
-
-Write mode introduces mutation risk (issue/PR changes, metadata updates). Risks include unauthorized action, accidental mass updates, and audit evasion.
-
-## Default Posture
-
-- `WRITE_MODE=false` by default.
-- When enabled, writes require repository whitelist membership by default.
-- Optional opt-in: `WRITE_ALLOW_ALL_TOKEN_REPOS=true` allows writes to any repo the token can access.
-- Policy engine remains authoritative and may deny specific write tools.
-
-## Supported Write Tools
-
-- `create_issue` (optional `milestone` id or title)
-- `update_issue` (optional `milestone`; `0` clears it)
-- `create_issue_comment`
-- `create_pr_comment`
-- `edit_issue_comment`
-- `add_labels`
-- `remove_labels`
-- `assign_issue`
-- `create_label`
-- `update_label`
-- `create_pull_request`
-- `create_release`
-- `edit_release`
-- `create_branch`
-- `create_milestone`
-
-Not supported (explicitly forbidden): merge actions, branch/label/release deletion,
-force push, repo/admin management, and repository content writes (file create/edit,
-commits). Gitea Projects (Kanban boards) are unsupported because the Gitea REST API
-exposes no project endpoints.
-
-## Enablement Steps
-
-1. Set `WRITE_MODE=true`.
-2. Choose one:
-   - `WRITE_REPOSITORY_WHITELIST=owner/repo,...` (recommended)
-   - `WRITE_ALLOW_ALL_TOKEN_REPOS=true` (broader scope)
-3. Review policy file for write-tool scope.
-4. Verify audit logging and alerting before rollout.
-
-## Safe Operations
-
-- Start with one repository in whitelist.
-- Use narrowly scoped bot credentials.
-- Require peer review for whitelist/policy changes.
-- Disable write mode during incident response if abuse is suspected.
-
-## Risk Tradeoffs
-
-Write mode improves automation and triage speed but increases blast radius. Use least privilege, tight policy, and strong monitoring.

scripts/check_docs.py

@@ -0,0 +1,414 @@
+"""Generate and check the tool inventory in the documentation.
+
+Usage:
+    uv run python scripts/check_docs.py          # check mode (CI job ``docs-check``)
+    uv run python scripts/check_docs.py --write  # regenerate docs/tools.md and the README block
+
+Check mode fails (exit code 1) when:
+
+* ``docs/tools.md`` differs from what the catalog generates;
+* the generated block of ``README.md`` (between ``<!-- tool-summary:start -->`` and
+  ``<!-- tool-summary:end -->``) differs from what the catalog generates;
+* ``README.md`` or a top-level ``docs/*.md`` file (``docs/adr``, ``docs/research`` and
+  ``docs/audit.md`` are excluded) mentions a ``gitea_*`` tool name that is not registered.
+
+The inventory comes from ``load_catalog()`` plus the hand-written raw tool
+``gitea_api_request``. Output is deterministic. The logic is importable so tests can
+drive it against a temporary tree.
+"""
+
+from __future__ import annotations
+
+import argparse
+import difflib
+import re
+import sys
+from collections import Counter
+from collections.abc import Iterable, Sequence
+from dataclasses import dataclass
+from pathlib import Path
+
+REPO_ROOT = Path(__file__).resolve().parents[1]
+sys.path.insert(0, str(REPO_ROOT / "src"))
+
+from aegis_gitea_mcp.catalog import Catalog, Operation, load_catalog  # noqa: E402
+from aegis_gitea_mcp.gitea import RAW_TOOL_NAME  # noqa: E402
+
+README_NAME = "README.md"
+TOOLS_DOC = Path("docs") / "tools.md"
+BLOCK_START = "<!-- tool-summary:start -->"
+BLOCK_END = "<!-- tool-summary:end -->"
+#: Documentation that may name tools that no longer exist (history, research, audit).
+EXCLUDED_DOCS = frozenset({"audit.md"})
+REGEN_HINT = "run: uv run python scripts/check_docs.py --write"
+
+RAW_DOMAIN = "misc"
+RAW_SUMMARY = "Raw request to any Gitea endpoint, for endpoints newer than the pinned spec"
+
+_TOOL_MENTION = re.compile(r"(?<![A-Za-z0-9_])gitea_[a-z0-9_]+")
+_MAX_DIFF_LINES = 24
+
+DOMAIN_BLURBS: dict[str, str] = {
+    "actions": (
+        "Gitea Actions: workflow runs, jobs, logs, artifacts, dispatch, secrets, variables "
+        "and runners at repository, organization, user and instance level."
+    ),
+    "admin": (
+        "Instance administration: users, organizations, emails, cron tasks, system hooks, "
+        "unadopted repositories, runners and packages. Needs a site-admin token."
+    ),
+    "hook": "Webhooks of repositories, organizations, users and the instance (system hooks).",
+    "issue": (
+        "Issues, comments, labels, milestones, reactions, pins, subscriptions, time tracking, "
+        "stopwatches, dependencies and attachments."
+    ),
+    "misc": (
+        "Version, settings, markdown rendering, license, gitignore and label templates, "
+        "signing keys, the current token, and the raw request tool."
+    ),
+    "notification": "Notification threads and their read state.",
+    "org": "Organizations, members, public members, blocks, labels, avatars and activity.",
+    "package": "Package registry: list, get, delete, files, link and unlink.",
+    "pull": "Pull requests, merging, reviews, review comments, requested reviewers, diffs.",
+    "release": "Releases, tags of releases and release attachments.",
+    "repo": (
+        "Repositories: CRUD, contents, branches, tags, commits, forks, mirrors, collaborators, "
+        "branch protection, deploy keys, archives, topics, LFS and more."
+    ),
+    "team": "Organization teams, their members and repositories.",
+    "user": (
+        "The signed-in user and other users: profile, emails, keys, followers, blocks, tokens, "
+        "OAuth applications, stars, watches, settings."
+    ),
+    "wiki": "Repository wiki pages and revisions.",
+}
+
+
+@dataclass(frozen=True)
+class ToolRow:
+    """One row of the inventory (a catalog operation or the raw request tool)."""
+
+    name: str
+    domain: str
+    method: str
+    path: str
+    summary: str
+    flags: tuple[str, ...]
+
+
+def _flags(operation: Operation) -> tuple[str, ...]:
+    flags: list[str] = []
+    if operation.read_only:
+        flags.append("read-only")
+    if operation.destructive:
+        flags.append("destructive")
+    if operation.paginated:
+        flags.append("paginated")
+    if any(parameter.is_file for parameter in operation.parameters):
+        flags.append("file upload")
+    if operation.domain == "admin":
+        flags.append("admin")
+    return tuple(flags)
+
+
+def build_rows(catalog: Catalog) -> list[ToolRow]:
+    """Return every tool (catalog operations plus the raw tool), sorted by domain then name."""
+    rows = [
+        ToolRow(
+            name=op.tool_name,
+            domain=op.domain,
+            method=op.method,
+            path=op.path,
+            summary=op.summary,
+            flags=_flags(op),
+        )
+        for op in catalog.operations
+    ]
+    rows.append(
+        ToolRow(
+            name=RAW_TOOL_NAME,
+            domain=RAW_DOMAIN,
+            method="any",
+            path="(caller supplied)",
+            summary=RAW_SUMMARY,
+            flags=("destructive",),
+        )
+    )
+    return sorted(rows, key=lambda row: (row.domain, row.name))
+
+
+def registered_names(catalog: Catalog) -> frozenset[str]:
+    """Return every registered tool name, including the raw tool."""
+    return frozenset({op.tool_name for op in catalog.operations} | {RAW_TOOL_NAME})
+
+
+def _cell(text: str) -> str:
+    """Make ``text`` safe inside a Markdown table cell."""
+    return " ".join(text.split()).replace("|", "\\|")
+
+
+def _domain_order(catalog: Catalog, rows: Iterable[ToolRow]) -> list[str]:
+    present = {row.domain for row in rows}
+    ordered = [domain for domain in catalog.domains() if domain in present]
+    ordered.extend(sorted(present - set(ordered)))
+    return ordered
+
+
+def render_summary_table(catalog: Catalog) -> str:
+    """Render the per-domain count table (also used inside the README block)."""
+    rows = build_rows(catalog)
+    totals: Counter[str] = Counter(row.domain for row in rows)
+    read_only: Counter[str] = Counter(row.domain for row in rows if "read-only" in row.flags)
+    lines = ["| Domain | Tools | Read-only | Mutating |", "|---|---:|---:|---:|"]
+    for domain in _domain_order(catalog, rows):
+        total = totals[domain]
+        lines.append(f"| {domain} | {total} | {read_only[domain]} | {total - read_only[domain]} |")
+    total_all = len(rows)
+    read_all = sum(read_only.values())
+    lines.append(f"| **all** | **{total_all}** | **{read_all}** | **{total_all - read_all}** |")
+    return "\n".join(lines)
+
+
+def render_readme_block(catalog: Catalog) -> str:
+    """Render the text that sits between the README markers (without the markers)."""
+    rows = build_rows(catalog)
+    generated = len(rows) - 1
+    body = (
+        f"{len(rows)} tools: {generated} generated from the Gitea {catalog.spec_version} "
+        f"swagger (one per operation) plus `{RAW_TOOL_NAME}`. The raw tool is counted "
+        "under misc.\n\n"
+        f"{render_summary_table(catalog)}\n\n"
+        "The full list with HTTP method, path and flags is in [docs/tools.md](docs/tools.md)."
+    )
+    return f"\n{body}\n"
+
+
+def render_tools_md(catalog: Catalog) -> str:
+    """Render ``docs/tools.md`` completely."""
+    rows = build_rows(catalog)
+    generated = len(rows) - 1
+    parts: list[str] = [
+        "# Tool inventory\n",
+        "This file is generated. Do not edit it by hand: run "
+        "`uv run python scripts/check_docs.py --write` after changing the catalog. CI fails "
+        "when it is out of date.\n",
+        f"There are {len(rows)} tools: {generated} generated from the pinned Gitea "
+        f"{catalog.spec_version} swagger (`spec/gitea-swagger.v1.json`, one tool per "
+        f"operation) plus the hand-written `{RAW_TOOL_NAME}`, which is listed under misc.\n",
+        "Naming: `gitea_<domain>_<action>`, lowercase snake_case, at most 64 characters. The "
+        "domain comes from the request path, the action from the swagger operationId. Names "
+        "are stable for a given spec version.\n",
+        "Every tool also accepts an optional `sudo` argument (a Gitea login). With an admin "
+        "token the call then runs as that user. Paths are relative to `/api/v1`.\n",
+        "Filters: `TOOLSETS` limits the exposed domains, `READ_ONLY` exposes only read-only "
+        f"operations (`{RAW_TOOL_NAME}` then accepts only GET), `DISABLED_TOOLS` hides "
+        "single tools by name. A filtered tool is neither listed nor callable. See "
+        "[configuration.md](configuration.md).\n",
+        "Flags: read-only (maps to the MCP `readOnlyHint`), destructive (`destructiveHint`: "
+        "deletes and calls that replace state), paginated (accepts `page` and `limit`), file "
+        "upload (takes base64 file content), admin (needs a site-admin token).\n",
+        "## Summary\n",
+        render_summary_table(catalog) + "\n",
+    ]
+    by_domain: dict[str, list[ToolRow]] = {}
+    for row in rows:
+        by_domain.setdefault(row.domain, []).append(row)
+    for domain in _domain_order(catalog, rows):
+        domain_rows = by_domain[domain]
+        parts.append(f"## {domain}\n")
+        blurb = DOMAIN_BLURBS.get(domain)
+        if blurb:
+            parts.append(f"{blurb} {len(domain_rows)} tools.\n")
+        table = [
+            "| Tool | Method | Path | Summary | Flags |",
+            "|---|---|---|---|---|",
+        ]
+        for row in domain_rows:
+            table.append(
+                f"| `{row.name}` | {row.method} | `{_cell(row.path)}` | {_cell(row.summary)} "
+                f"| {', '.join(row.flags)} |"
+            )
+        parts.append("\n".join(table) + "\n")
+    return "\n".join(parts)
+
+
+# ---------------------------------------------------------------------------
+# README block handling
+# ---------------------------------------------------------------------------
+
+
+def extract_block(readme_text: str) -> str | None:
+    """Return the text between the README markers, or None when a marker is missing."""
+    start = readme_text.find(BLOCK_START)
+    end = readme_text.find(BLOCK_END)
+    if start == -1 or end == -1 or end < start:
+        return None
+    return readme_text[start + len(BLOCK_START) : end]
+
+
+def replace_block(readme_text: str, block: str) -> str:
+    """Return ``readme_text`` with the generated block replaced (markers must exist)."""
+    start = readme_text.find(BLOCK_START)
+    end = readme_text.find(BLOCK_END)
+    if start == -1 or end == -1 or end < start:
+        raise ValueError(f"README.md must contain {BLOCK_START} and {BLOCK_END} (in that order)")
+    return readme_text[: start + len(BLOCK_START)] + block + readme_text[end:]
+
+
+# ---------------------------------------------------------------------------
+# Tool name mentions
+# ---------------------------------------------------------------------------
+
+
+def doc_files(root: Path) -> list[Path]:
+    """Return the Markdown files scanned for tool names (README plus top-level docs)."""
+    files: list[Path] = []
+    readme = root / README_NAME
+    if readme.is_file():
+        files.append(readme)
+    docs = root / "docs"
+    if docs.is_dir():
+        files.extend(
+            path
+            for path in sorted(docs.glob("*.md"))
+            if path.name not in EXCLUDED_DOCS and path.is_file()
+        )
+    return files
+
+
+def is_known_mention(mention: str, names: frozenset[str]) -> bool:
+    """Return True for a registered tool name or a prefix such as ``gitea_admin_``."""
+    if mention in names:
+        return True
+    return mention.endswith("_") and any(name.startswith(mention) for name in names)
+
+
+def find_unknown_mentions(text: str, names: frozenset[str]) -> list[tuple[int, str]]:
+    """Return ``(line number, name)`` for every mention that is not a registered tool."""
+    unknown: list[tuple[int, str]] = []
+    for number, line in enumerate(text.splitlines(), start=1):
+        for match in _TOOL_MENTION.finditer(line):
+            mention = match.group(0)
+            if not is_known_mention(mention, names):
+                unknown.append((number, mention))
+    return unknown
+
+
+# ---------------------------------------------------------------------------
+# check and write
+# ---------------------------------------------------------------------------
+
+
+def _diff_summary(label: str, current: str, expected: str) -> str:
+    diff = list(
+        difflib.unified_diff(
+            current.splitlines(),
+            expected.splitlines(),
+            fromfile=f"{label} (committed)",
+            tofile=f"{label} (generated)",
+            lineterm="",
+            n=0,
+        )
+    )
+    shown = diff[:_MAX_DIFF_LINES]
+    if len(diff) > len(shown):
+        shown.append(f"... {len(diff) - len(shown)} more diff lines")
+    return "\n".join(shown)
+
+
+def check(root: Path, catalog: Catalog | None = None) -> list[str]:
+    """Run every drift rule against the tree at ``root`` and return the problems found."""
+    catalog = catalog or load_catalog()
+    problems: list[str] = []
+
+    tools_path = root / TOOLS_DOC
+    expected_tools = render_tools_md(catalog)
+    if not tools_path.is_file():
+        problems.append(f"{TOOLS_DOC} is missing. {REGEN_HINT}")
+    else:
+        current = tools_path.read_text(encoding="utf-8")
+        if current != expected_tools:
+            problems.append(
+                f"{TOOLS_DOC} is out of date. {REGEN_HINT}\n"
+                + _diff_summary(str(TOOLS_DOC), current, expected_tools)
+            )
+
+    readme_path = root / README_NAME
+    if not readme_path.is_file():
+        problems.append(f"{README_NAME} is missing")
+    else:
+        readme = readme_path.read_text(encoding="utf-8")
+        block = extract_block(readme)
+        expected_block = render_readme_block(catalog)
+        if block is None:
+            problems.append(
+                f"{README_NAME} has no generated block: add {BLOCK_START} and {BLOCK_END}, "
+                "then run: uv run python scripts/check_docs.py --write"
+            )
+        elif block != expected_block:
+            problems.append(
+                f"the tool summary in {README_NAME} is out of date. {REGEN_HINT}\n"
+                + _diff_summary(f"{README_NAME} tool summary", block, expected_block)
+            )
+
+    names = registered_names(catalog)
+    for path in doc_files(root):
+        text = path.read_text(encoding="utf-8")
+        for number, mention in find_unknown_mentions(text, names):
+            problems.append(
+                f"{path.relative_to(root)}:{number}: '{mention}' is not a registered tool name"
+            )
+    return problems
+
+
+def write(root: Path, catalog: Catalog | None = None) -> list[Path]:
+    """Regenerate ``docs/tools.md`` and the README block. Returns the files written."""
+    catalog = catalog or load_catalog()
+    written: list[Path] = []
+
+    tools_path = root / TOOLS_DOC
+    tools_path.parent.mkdir(parents=True, exist_ok=True)
+    tools_path.write_text(render_tools_md(catalog), encoding="utf-8")
+    written.append(tools_path)
+
+    readme_path = root / README_NAME
+    readme = readme_path.read_text(encoding="utf-8")
+    updated = replace_block(readme, render_readme_block(catalog))
+    if updated != readme:
+        readme_path.write_text(updated, encoding="utf-8")
+    written.append(readme_path)
+    return written
+
+
+def main(argv: Sequence[str] | None = None, root: Path = REPO_ROOT) -> int:
+    """Run the script and return the process exit code."""
+    parser = argparse.ArgumentParser(description=__doc__.splitlines()[0])
+    parser.add_argument(
+        "--write",
+        action="store_true",
+        help="regenerate docs/tools.md and the README tool summary instead of checking",
+    )
+    args = parser.parse_args(argv)
+
+    if args.write:
+        try:
+            written = write(root)
+        except ValueError as exc:
+            print(f"error: {exc}", file=sys.stderr)
+            return 1
+        for path in written:
+            print(f"wrote {path.relative_to(root)}")
+        return 0
+
+    problems = check(root)
+    if problems:
+        print(f"docs check failed with {len(problems)} problem(s):", file=sys.stderr)
+        for problem in problems:
+            print(f"- {problem}", file=sys.stderr)
+        return 1
+    print("docs check passed")
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

tests/test_check_docs.py

@@ -0,0 +1,239 @@
+"""Tests for scripts/check_docs.py: generation, drift detection and tool name mentions."""
+
+from __future__ import annotations
+
+import importlib.util
+import re
+import sys
+from pathlib import Path
+from types import ModuleType
+
+import pytest
+
+from aegis_gitea_mcp.catalog import load_catalog
+
+REPO_ROOT = Path(__file__).resolve().parents[1]
+SCRIPT = REPO_ROOT / "scripts" / "check_docs.py"
+
+README_TEMPLATE = (
+    "# Title\n\nIntro.\n\n<!-- tool-summary:start -->\nold\n<!-- tool-summary:end -->\n\nOutro.\n"
+)
+
+
+@pytest.fixture(scope="module")
+def script() -> ModuleType:
+    spec = importlib.util.spec_from_file_location("check_docs_script", SCRIPT)
+    assert spec is not None
+    assert spec.loader is not None
+    module = importlib.util.module_from_spec(spec)
+    sys.modules[spec.name] = module  # dataclasses resolves string annotations through it
+    spec.loader.exec_module(module)
+    return module
+
+
+@pytest.fixture
+def tree(script: ModuleType, tmp_path: Path) -> Path:
+    """A minimal docs tree in which the generated files are up to date."""
+    (tmp_path / "docs").mkdir()
+    (tmp_path / "README.md").write_text(README_TEMPLATE, encoding="utf-8")
+    script.write(tmp_path)
+    return tmp_path
+
+
+def test_repository_docs_are_in_sync(script: ModuleType) -> None:
+    # docs/security.md still describes the 0.x design until the security phase rewrites it;
+    # its stale tool names are the only accepted problems. Drop the filter after that rewrite.
+    problems = [p for p in script.check(REPO_ROOT) if not p.startswith("docs/security.md:")]
+    assert problems == []
+
+
+def test_tools_md_lists_every_tool_once(script: ModuleType) -> None:
+    catalog = load_catalog()
+    text = script.render_tools_md(catalog)
+    rows = re.findall(r"^\| `(gitea_[a-z0-9_]+)` \|", text, flags=re.MULTILINE)
+    expected = {op.tool_name for op in catalog.operations} | {"gitea_api_request"}
+    assert len(rows) == len(expected) == len(catalog.operations) + 1
+    assert set(rows) == expected
+    assert f"There are {len(expected)} tools" in text
+    assert catalog.spec_version in text
+
+
+def test_tools_md_row_content(script: ModuleType) -> None:
+    catalog = load_catalog()
+    text = script.render_tools_md(catalog)
+    delete = next(op for op in catalog.operations if op.tool_name == "gitea_repo_delete")
+    row = next(line for line in text.splitlines() if line.startswith("| `gitea_repo_delete` |"))
+    assert f"| {delete.method} | `{delete.path}` |" in row
+    assert "destructive" in row
+    raw_row = next(line for line in text.splitlines() if line.startswith("| `gitea_api_request`"))
+    assert "any" in raw_row
+    assert "destructive" in raw_row
+
+
+def test_flags_cover_admin_upload_and_pagination(script: ModuleType) -> None:
+    catalog = load_catalog()
+    rows = {row.name: row for row in script.build_rows(catalog)}
+    admin = next(op.tool_name for op in catalog.operations if op.domain == "admin")
+    assert "admin" in rows[admin].flags
+    assert any("file upload" in row.flags for row in rows.values())
+    assert any("paginated" in row.flags for row in rows.values())
+    assert any("read-only" in row.flags for row in rows.values())
+
+
+def test_output_is_deterministic(script: ModuleType) -> None:
+    catalog = load_catalog()
+    assert script.render_tools_md(catalog) == script.render_tools_md(catalog)
+    assert script.render_readme_block(catalog) == script.render_readme_block(catalog)
+
+
+def test_summary_table_counts_match_catalog(script: ModuleType) -> None:
+    catalog = load_catalog()
+    block = script.render_readme_block(catalog)
+    total = len(catalog.operations) + 1
+    assert f"{total} tools" in block
+    misc = sum(1 for op in catalog.operations if op.domain == "misc") + 1
+    assert f"| misc | {misc} |" in block
+    assert f"| **all** | **{total}** |" in block
+
+
+def test_table_cells_escape_pipes_and_newlines(script: ModuleType) -> None:
+    assert script._cell("a | b\nc") == "a \\| b c"
+
+
+def test_write_mode_creates_files_and_replaces_only_the_block(
+    script: ModuleType, tmp_path: Path
+) -> None:
+    (tmp_path / "README.md").write_text(README_TEMPLATE, encoding="utf-8")
+    written = script.write(tmp_path)
+    assert {path.name for path in written} == {"tools.md", "README.md"}
+    readme = (tmp_path / "README.md").read_text(encoding="utf-8")
+    assert readme.startswith("# Title\n\nIntro.\n\n<!-- tool-summary:start -->")
+    assert readme.endswith("<!-- tool-summary:end -->\n\nOutro.\n")
+    assert "old" not in readme
+    assert (tmp_path / "docs" / "tools.md").read_text(encoding="utf-8").startswith("# Tool")
+
+
+def test_write_mode_is_idempotent(script: ModuleType, tree: Path) -> None:
+    before = (tree / "README.md").read_text(encoding="utf-8")
+    script.write(tree)
+    assert (tree / "README.md").read_text(encoding="utf-8") == before
+
+
+def test_write_requires_markers(script: ModuleType, tmp_path: Path) -> None:
+    (tmp_path / "README.md").write_text("# no markers\n", encoding="utf-8")
+    with pytest.raises(ValueError, match="tool-summary"):
+        script.write(tmp_path)
+    assert script.main(["--write"], root=tmp_path) == 1
+
+
+def test_main_write_then_check(
+    script: ModuleType, tmp_path: Path, capsys: pytest.CaptureFixture[str]
+) -> None:
+    (tmp_path / "README.md").write_text(README_TEMPLATE, encoding="utf-8")
+    assert script.main(["--write"], root=tmp_path) == 0
+    assert "wrote docs/tools.md" in capsys.readouterr().out
+    assert script.main([], root=tmp_path) == 0
+    assert "docs check passed" in capsys.readouterr().out
+
+
+def test_clean_tree_passes(script: ModuleType, tree: Path) -> None:
+    assert script.check(tree) == []
+
+
+def test_detects_stale_tools_md(
+    script: ModuleType, tree: Path, capsys: pytest.CaptureFixture[str]
+) -> None:
+    path = tree / "docs" / "tools.md"
+    path.write_text(path.read_text(encoding="utf-8").replace("| repo |", "| repos |"))
+    problems = script.check(tree)
+    assert len(problems) == 1
+    assert "docs/tools.md is out of date" in problems[0]
+    assert "check_docs.py --write" in problems[0]
+    assert "-| repos |" in problems[0]
+    assert "+| repo |" in problems[0]
+    assert script.main([], root=tree) == 1
+    assert "docs check failed" in capsys.readouterr().err
+
+
+def test_detects_missing_tools_md(script: ModuleType, tree: Path) -> None:
+    (tree / "docs" / "tools.md").unlink()
+    problems = script.check(tree)
+    assert any("docs/tools.md is missing" in problem for problem in problems)
+
+
+def test_detects_stale_readme_block(script: ModuleType, tree: Path) -> None:
+    readme = tree / "README.md"
+    readme.write_text(readme.read_text(encoding="utf-8").replace("| misc |", "| miscx |"))
+    problems = script.check(tree)
+    assert len(problems) == 1
+    assert "tool summary in README.md is out of date" in problems[0]
+
+
+def test_detects_missing_readme_markers(script: ModuleType, tree: Path) -> None:
+    (tree / "README.md").write_text("# no block here\n", encoding="utf-8")
+    problems = script.check(tree)
+    assert any("no generated block" in problem for problem in problems)
+
+
+def test_detects_reversed_markers(script: ModuleType, tree: Path) -> None:
+    (tree / "README.md").write_text(
+        "<!-- tool-summary:end -->\n<!-- tool-summary:start -->\n", encoding="utf-8"
+    )
+    assert any("no generated block" in problem for problem in script.check(tree))
+
+
+def test_detects_unknown_tool_name_in_readme(script: ModuleType, tree: Path) -> None:
+    readme = tree / "README.md"
+    readme.write_text(readme.read_text(encoding="utf-8") + "\nUse `gitea_made_up_tool`.\n")
+    problems = script.check(tree)
+    assert len(problems) == 1
+    assert "README.md" in problems[0]
+    assert "'gitea_made_up_tool' is not a registered tool name" in problems[0]
+
+
+def test_detects_unknown_tool_name_in_docs(script: ModuleType, tree: Path) -> None:
+    (tree / "docs" / "guide.md").write_text("one\ntwo `gitea_request` three\n", encoding="utf-8")
+    problems = script.check(tree)
+    assert problems == ["docs/guide.md:2: 'gitea_request' is not a registered tool name"]
+
+
+def test_registered_names_and_prefixes_are_accepted(script: ModuleType, tree: Path) -> None:
+    (tree / "docs" / "guide.md").write_text(
+        "`gitea_repo_get`, `gitea_api_request` and the gitea_admin_ tools.\n", encoding="utf-8"
+    )
+    assert script.check(tree) == []
+
+
+def test_unknown_prefix_is_rejected(script: ModuleType, tree: Path) -> None:
+    (tree / "docs" / "guide.md").write_text("the gitea_bogus_ tools\n", encoding="utf-8")
+    assert len(script.check(tree)) == 1
+
+
+def test_similar_identifiers_are_not_tool_mentions(script: ModuleType) -> None:
+    names = script.registered_names(load_catalog())
+    text = "aegis_gitea_mcp, my_gitea_thing, GITEA_URL and gitea_<domain>_<action>\n"
+    assert script.find_unknown_mentions(text, names) == []
+
+
+def test_excluded_documents_may_name_old_tools(script: ModuleType, tree: Path) -> None:
+    (tree / "docs" / "audit.md").write_text("old `gitea_request`\n", encoding="utf-8")
+    (tree / "docs" / "adr").mkdir()
+    (tree / "docs" / "adr" / "0001.md").write_text("old `gitea_request`\n", encoding="utf-8")
+    (tree / "docs" / "research").mkdir()
+    (tree / "docs" / "research" / "notes.md").write_text("`gitea_request`\n", encoding="utf-8")
+    assert script.check(tree) == []
+    scanned = {path.name for path in script.doc_files(tree)}
+    assert scanned == {"README.md", "tools.md"}
+
+
+def test_missing_readme_is_reported(script: ModuleType, tree: Path) -> None:
+    (tree / "README.md").unlink()
+    assert any("README.md is missing" in problem for problem in script.check(tree))
+
+
+def test_long_diffs_are_truncated(script: ModuleType) -> None:
+    current = "\n".join(f"old {i}" for i in range(200))
+    expected = "\n".join(f"new {i}" for i in range(200))
+    summary = script._diff_summary("x", current, expected)
+    assert "more diff lines" in summary
+    assert len(summary.splitlines()) <= 30