skills/clawlink-hq/skills/fathom/Gen Agent Trust Hub

fathom

Pass

Audited by Gen Agent Trust Hub on Sep 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses npx @useclawlink/cli to perform authentication and execute actions in Fathom. These commands are part of the core functionality for the vendor's integration.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes npx to download and execute the @useclawlink/cli package from the NPM registry. This is the official tool provided by the vendor for interacting with their service.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data fetched from Fathom meetings and recordings. While this introduces a surface for indirect prompt injection where meeting notes might contain instructions for the agent, the risk is mitigated by the instruction to 'confirm with the user before any write' and standard agent guardrails.
  • Ingestion points: Data returned from npx @useclawlink/cli run fathom (e.g., meeting transcripts, summaries).
  • Boundary markers: Explicitly instructs the agent to read data first and confirm with the user before performing any write operations.
  • Capability inventory: The skill can perform network operations via the CLI and access the local file system at ~/.clawlink/ to store credentials.
  • Sanitization: Relies on the agent's internal content filtering and the required user confirmation for writes.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 23, 2026, 05:45 PM