setup-benny
Pass
Audited by Gen Agent Trust Hub on Nov 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow for creating automations that process untrusted external data, which could be used to influence the agent's behavior.
- Ingestion points: The triage and reproduction automations ingest content from Slack channel reports and threads (Section 7).
- Boundary markers: The reproduction automation is instructed to wait for a 'trusted triage marker' before acting (Section 7), but there are no defined delimiters or instructions to ignore embedded commands within the Slack report content itself.
- Capability inventory: The configured automations possess significant capabilities, including reading and replying to Slack threads, performing CRUD operations on issue trackers, accessing repository history, and creating draft Pull Requests (Sections 4 and 7).
- Sanitization: The skill lacks explicit instructions for sanitizing or escaping the data retrieved from Slack before it is interpolated into agent prompts.
Audit Metadata