encode-skill

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests data from external API endpoints, which creates an attack surface for indirect prompt injection if the response contains malicious instructions.\n
  • Ingestion points: Data is ingested through the execute function in scripts/rest_request.py via HTTP responses.\n
  • Boundary markers: Absent; the skill does not wrap the external content in protective delimiters or provide instructions to ignore embedded commands.\n
  • Capability inventory: The skill has network access (via requests) and local file-writing capabilities (via path.write_text).\n
  • Sanitization: Basic truncation is performed on the output, but no security-focused filtering or sanitization of string content is applied.\n- [DATA_EXFILTRATION]: The helper script scripts/rest_request.py functions as a generic REST client that allows requests to any URL provided in the base_url parameter. This flexibility could be leveraged to send data to non-whitelisted or attacker-controlled domains.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 03:57 PM
Security Audit — agent-trust-hub — encode-skill