Skip to content

Agent Integration

SiteSkills skills are agent-agnostic. A skill is a set of structured Markdown instructions -- any agent that can read files and control a browser can use them.

Supported Agents

AgentTypeInstall PathNotes
Claude CoworkNative SKILL.md.claude/skills/Auto-discovers skills from directory
Browser UsePython framework./skills/Pass skills directory to agent config
StagehandTypeScript frameworkConfigurableRead SKILL.md for navigation guidance
Playwright MCPMCP serverConfigurableAny MCP-compatible agent can use skills

How It Works

  1. Install a skill using the CLI: siteskills install <skill-name>
  2. The CLI places the skill directory in the correct location for your agent
  3. The agent reads SKILL.md when a matching trigger phrase is detected
  4. The agent follows the navigation steps, using DOM hints and fallbacks
  5. Execution results can be reported back via siteskills report

Skill Discovery

Agents discover skills through different mechanisms:

  • Claude Cowork: Scans .claude/skills/ on startup, matches triggers from SKILL.md frontmatter
  • Browser Use: Loads from configured skills_dir, matches by domain and trigger
  • Stagehand: Skills are loaded programmatically or referenced in agent configuration

Custom Agent Integration

If you are building your own agent, here is how to consume SiteSkills skills:

  1. Download skills via the API or CLI
  2. Parse the SKILL.md frontmatter for name, description, and triggers
  3. When a user request matches a trigger, load the full SKILL.md
  4. Follow the Navigation Flow steps sequentially
  5. Use DOM hints to find elements (try selectors in priority order)
  6. Respect Wait condition timeouts before proceeding
  7. On failure, follow the Fallback instructions for each step
  8. After completion, check the Verification section to confirm success

Execution Reporting

Report execution results to improve community success rates:

bash
siteskills report <skill-name> --success --time 12400 --agent my-agent

Or via the API:

POST /api/telemetry/execution

See API Reference for the full telemetry payload.

Built for the AI browser agent ecosystem