docs-page

Fail

Audited by Gen Agent Trust Hub on May 9, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The example documentation file contains a shell command that pipes a remote script from an unverified domain (https://get.filebase.dev) directly into a shell (curl | sh). This pattern facilitates the execution of untrusted code and is a high-risk security practice.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted data from user briefs and external configuration files without sanitization.\n
  • Ingestion points: The skill reads an external DESIGN.md file and a user-provided topic/brief to generate its documentation output.\n
  • Boundary markers: There are no explicit delimiters or instructions to treat the ingested content strictly as data, potentially allowing embedded instructions to influence the agent.\n
  • Capability inventory: The skill generates complex HTML documents which could be used to host malicious scripts or phishing content if the input source is compromised.\n
  • Sanitization: The workflow does not specify any input validation, escaping, or filtering of the external data before it is interpolated into the final artifact.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
May 9, 2026, 04:29 AM
Security Audit — agent-trust-hub — docs-page