orca-linear
Fail
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: HIGHPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PRIVILEGE_ESCALATION]: The instructions explicitly direct the agent to re-run commands with escalated permissions if an access denied error occurs, which may lead the agent to use sudo or other administrative privileges.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from Linear tickets, comments, and attachments. This creates an attack surface where malicious instructions could be embedded in ticket data to influence the agent.
- Ingestion points: Linear ticket descriptions, comments, and file attachments mentioned in SKILL.md.
- Boundary markers: Includes a natural language warning in the description of SKILL.md to treat ticket data as untrusted.
- Capability inventory: Execution of system commands via the Orca CLI in SKILL.md.
- Sanitization: No automated sanitization or structured validation of ticket content is described in SKILL.md.
- [COMMAND_EXECUTION]: The skill executes a local binary (Orca) resolved through environment variables or specific naming conventions in SKILL.md. This involves running system-level commands to interact with external services.
Recommendations
- AI detected serious security threats
Audit Metadata