status-report
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external and potentially untrusted sources, creating a surface for indirect prompt injection.
- Ingestion points: The skill reads data from
ReadNotificationsand fetches Pull Request details using theghtool (GitHub CLI). - Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat the ingested data as untrusted or to ignore instructions embedded within the notifications or PR content.
- Capability inventory: While the skill instructions state that every task and PR stays as it is (read-only), the skill operates within a broader agent context that possesses tools for code modification and command execution.
- Sanitization: The instructions do not specify any sanitization, filtering, or escaping of the content retrieved from external sources before including it in the report output.
Audit Metadata