goal-framing
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of user-provided task descriptions which are then formatted into templates like TaskIntentDraft and SubagentContextPacket for further processing by the agent or subagents.
- Ingestion points: User-supplied task descriptions following the /aegis-goal or Aegis goal: triggers in SKILL.md.
- Boundary markers: The skill uses YAML-like key-value structures to delimit fields (e.g.,
- Goal:,
- Stop condition:), but lacks explicit instructions to treat the user content as untrusted data or to ignore embedded commands.
- Capability inventory: The skill does not have direct execution capabilities but directs the agent to other workflows (debugging, planning) and mentions writing persistent records to the docs/aegis/ directory.
- Sanitization: No sanitization, escaping, or validation instructions are provided for the user-supplied task descriptions.
Audit Metadata