aporia-design-process

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill follows security best practices for secret management by instructing users to configure API keys via environment variables rather than hardcoding credentials. It utilizes a specific MCP server (aporia) to scope its tool access to the intended service.
  • [INDIRECT_PROMPT_INJECTION]: The skill features a data ingestion surface by reading external context from the Aporia platform, including personas and existing process structures. While this represents a potential vector for indirect prompt injection, the risk is mitigated because the data is sourced from the user's own environment, and the skill mandates a visual validation step (Phase 2 Shape Gate) where the user must approve the drafted process before any changes are recorded.
  • Ingestion points: External data is ingested via aporia:pull_constitution and aporia:pull_context as described in Phase 0 and 0b of SKILL.md.
  • Boundary markers: No explicit delimiter markers are used for tool output data.
  • Capability inventory: The skill can create feature nodes (aporia:apply_scan), record processes (aporia:record_process), and record notes (aporia:record_notes).
  • Sanitization: The skill utilizes manual human review of an ASCII preview as a primary sanitization and validation mechanism before persistence.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 06:24 AM
Security Audit — agent-trust-hub — aporia-design-process