hawk-survey

Pass

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The hawk-survey skill performs a wide-ranging audit of project source code and infrastructure configuration. This process requires the agent to process large volumes of untrusted data from the local filesystem.
  • Ingestion points: The agent is instructed to read all files in the defined project scope using commands like find, grep, and cat during Phase 2 (DESCEND) and Phase 3 (ASSESS).
  • Boundary markers: The instructions do not define specific delimiters or security guardrails that direct the agent to ignore instructions embedded within the files it is analyzing, which could lead to the agent following malicious commands found in comments or documentation.
  • Capability inventory: The skill grants the agent capabilities to search the filesystem, execute auditing tools such as pnpm audit and npx svelte-check, and modify code during the remediation phase (Phase 5).
  • Sanitization: Content read from project files is incorporated directly into the agent's context without sanitization or filtering, creating a potential vector for an attacker to manipulate the security assessment result by placing injection patterns in the audited codebase.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 9, 2026, 01:31 AM