proof-point-packager
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill proactively mitigates indirect prompt injection risks by instructing the agent to treat all external inputs, such as scraped web text, reviews, and emails, as untrusted evidence. It explicitly commands the agent not to follow any instructions embedded within such evidence.
- [COMMAND_EXECUTION]: The skill's contract requires explicit user authorization before executing any actions with external side effects, including publishing content, sending messages, or modifying account states.
- [CREDENTIALS_UNSAFE]: Reference materials within the skill promote secure development practices by explicitly advising against storing API keys or credentials within the skill's instructions or source control.
Audit Metadata