good-strategy-bad-strategy
Pass
Audited by Gen Agent Trust Hub on Aug 23, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions in SKILL.md describe auditing user-provided strategy documents and pitch decks, which creates a surface for indirect prompt injection. Maliciously crafted instructions within these audited documents could potentially override the agent's behavior. 1) Ingestion points: Auditing strategy documents and pitch decks. 2) Boundary markers: Absent. 3) Capability inventory: Inherited from host agent. 4) Sanitization: Absent.
- [METADATA_POISONING]: The YAML frontmatter in SKILL.md identifies the author as 'wondelai', while the provided author context is 'manish31-oss'. This inconsistency in metadata can be deceptive regarding the provenance and source of the skill's instructions.
Audit Metadata