ss-proposal

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill utilizes the public PlantUML instance at https://www.plantuml.com/plantuml to render architectural and sequence diagrams. This involves encoding and transmitting diagram source descriptions—which may include sensitive architecture details, class structures, and data flows—to an external service.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from external sources during proposal generation.\n
  • Ingestion points: Processes content from external PRD links, requirement descriptions, and Figma design URLs (as specified in SKILL.md).\n
  • Boundary markers: No explicit delimiters or boundary markers are used to isolate untrusted input content from the agent's instructions.\n
  • Capability inventory: Accesses project files (codebases, documentation), performs network requests to fetch external documents, writes Markdown files to docs/proposals/, and can spawn independent subagents for review tasks.\n
  • Sanitization: No input validation or sanitization of the content fetched from external documents is required or performed.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 09:08 AM
Security Audit — agent-trust-hub — ss-proposal