Coding agents
Run squirrelscan audits from Claude Code, Cursor, Codex, ChatGPT, opencode, and any MCP client: setup, MCP, skills, and the fix loop.
squirrelscan is built for coding agents. Point your agent at squirrelscan and it can audit a site, read the findings as structured tool output, and fix them in your codebase, without you copy-pasting reports around.
Pick your agent
Plugin (skills + MCP in one), or claude mcp add.
One-click install, or a two-line mcp.json.
codex mcp add or ~/.codex/config.toml.
Add squirrelscan as a developer-mode connector.
Remote MCP in opencode.json.
VS Code, Windsurf, Gemini CLI, and the raw endpoint.
Three ways to connect
Every agent uses one or more of these. The per-agent pages above show the exact steps.
The deepest integration. The hosted server at mcp.squirrelscan.com exposes audits, rules, issues, and reports as native tools, no shell piping. See Hosted MCP.
Autonomous audit + fix workflows the agent discovers and runs on its own. Install once: npx skills add squirrelscan/squirrelscan.
No integration needed: squirrel audit example.com --format llm | <agent>. Works with anything that reads stdin.
The skills
The Agent Skills standard ships two squirrelscan skills. They install the same way for every agent (npx skills add squirrelscan/squirrelscan, landing in .agents/skills/); Claude Code can also get them bundled in the plugin.
| Skill | What it does |
|---|---|
squirrelscan | Operate the CLI: install, login, run audits, publish reports, credits, API keys, MCP setup, configuration |
audit-website | The fix loop: audit a site, map findings to code, fix in batches, re-audit until it scores well |
The fix loop
Once connected, the pattern is the same everywhere. Ask the agent to audit and fix:
Use the audit-website skill to audit example.com and fix the top issues.
The agent will:
- Run the audit (MCP tool or CLI).
- Group findings by severity and category.
- Map each issue to the source file.
- Apply fixes in batches.
- Re-audit to confirm the score improved.
Configuration for agents
Drop a squirrel.toml in your project so every agent run uses consistent settings:
[crawler]
max_pages = 50
respect_robots = false # default: audits are owner-initiated
[rules]
disable = ["content/word-count", "content/reading-level"]See the configuration reference for all options, and Fix your site with an AI agent for a full worked example.
Related
- Hosted MCP server - the tool surface, auth, and how credits work
- MCP client setup - copy-paste config for every client
squirrelCLI - the command line the skills wrap