setup-serena-mcp

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill retrieves documentation and setup instructions from external URLs, specifically https://raw.githubusercontent.com/oraios/serena/refs/heads/main/README.md and multiple pages under https://oraios.github.io/serena/.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external content from documentation URLs to guide the agent's setup process, creating a risk that malicious instructions in the source documents could influence the agent's actions.
  • Ingestion points: External documentation URLs listed in Step 3 of SKILL.md are loaded into the agent's context.
  • Boundary markers: The skill lacks delimiters or explicit instructions to ignore potentially conflicting or malicious directions embedded in the fetched documentation.
  • Capability inventory: The skill has the capability to write to local and user-level configuration files (CLAUDE.md, CLAUDE.local.md, ~/.claude/CLAUDE.md) and guides the execution of CLI tools like uv and serena.
  • Sanitization: No sanitization or verification of the external content is performed before it is used to guide the installation workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 01:10 AM
Security Audit — agent-trust-hub — setup-serena-mcp