prowlarr
Pass
Audited by Gen Agent Trust Hub on Apr 19, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The
SKILL.mdfile uses coercive and authoritative language ('MANDATORY SKILL INVOCATION', 'YOU MUST', 'Failure ... violates your operational requirements') to override the agent's decision-making logic regarding tool use. - [PROMPT_INJECTION]: The skill presents untrusted data from external Prowlarr indexers to the agent without sanitization, exposing it to indirect prompt injection.
- Ingestion points: External search results retrieved via
scripts/prowlarr-api.sh. - Boundary markers: None; external data is presented directly to the agent's context.
- Capability inventory: Shell command execution (
curl,jq) viazsh-tool. - Sanitization: None; data is processed for formatting but field contents are not validated for instructions.
- [COMMAND_EXECUTION]: The script
scripts/prowlarr-api.shexecutes bash commands includingcurlandjqto interact with the local Prowlarr service.
Audit Metadata