agentation
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs the
agentationpackage from the npm registry to provide the toolbar component. - [COMMAND_EXECUTION]: The setup process involves executing several commands to manage the environment and tools:
npm install agentationfor package installation.claude mcp add agentation -- npx agentation-mcp serverto register the Model Context Protocol server for persistent tool access.npx agentation-mcp doctorto perform a diagnostic check on the setup.- [INDIRECT_PROMPT_INJECTION]: The skill facilitates syncing annotations from a visual web toolbar into the agent's context, which introduces a potential attack surface.
- Ingestion points: Annotations are collected from the web interface and processed by the agent through the
agentation-mcpserver (SKILL.md). - Boundary markers: The instructions do not define specific delimiters to isolate the synced data from the agent's primary instructions.
- Capability inventory: The skill allows the agent to modify project layout files (
layout.tsx,_app.tsx) and execute shell commands for setup and verification (SKILL.md). - Sanitization: No evidence of sanitization or filtering for the external annotation content is provided in the skill documentation.
Audit Metadata