linearis
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFECREDENTIALS_UNSAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill accesses the sensitive configuration file
~/.config/catalyst-cloud/customer.jsonto retrieve the API key used for authentication. While necessary for the skill's functionality, this introduces a risk of credential exposure if the agent is subverted. - [COMMAND_EXECUTION]: The skill performs shell commands using
curlto interact with a mirror API. ThebaseUrland authentication headers are derived from local configuration, which could be exploited to send data to unintended endpoints. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from ticket titles, bodies, and comments which can contain malicious instructions.
- Ingestion points: Ticket data and comments are ingested via API requests described in
SKILL.md. - Boundary markers: No specific boundary markers or instructions to ignore embedded commands are implemented when processing ticket content.
- Capability inventory: The skill possesses shell execution capabilities (
curl) and read access to local credential files (~/.config/catalyst-cloud/customer.json). - Sanitization: No sanitization of ticket content is mentioned before it is processed by the agent.
Audit Metadata