bid-proposal-writer

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses curl commands to manage files and trigger tasks on a secondary service. This behavior is transparently documented and serves the primary function of the skill.
  • [EXTERNAL_DOWNLOADS]: The skill communicates with an external endpoint (http://opencode-bridge:8000). This service name suggests it is an internal or infrastructure-specific component used for processing user-uploaded documents.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external documents (RFP/bid materials).
  • Ingestion points: Files uploaded by the user to the input/ directory, which are then read and sent to the bridge service.
  • Boundary markers: None provided in the instruction templates; the agent is instructed to process the documents directly.
  • Capability inventory: File reading (input/), network communication via curl, and file writing to the output/ directory.
  • Sanitization: There are no explicit instructions for the agent to sanitize or ignore embedded instructions within the processed documents.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 09:45 AM
Security Audit — agent-trust-hub — bid-proposal-writer