paywalls
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill handles untrusted external data, such as scraped text, customer reviews, and emails. It proactively mitigates the risk of indirect prompt injection by including specific instructions in
SKILL.mdandreferences/skill-contract.mdto treat these sources as evidence only and to ignore any instructions embedded within them. - [COMMAND_EXECUTION]: While the skill and its reference files (e.g.,
references/connectors.md) mention interaction with marketing execution tools and deterministic scripts, they enforce a strict policy requiring explicit user authorization for any persistent writes, account changes, or real-world side effects. - [CREDENTIALS_UNSAFE]: The skill follows security best practices by explicitly instructing developers and agents to keep API keys out of skills and source control, as documented in
references/connectors.md.
Audit Metadata