orca-cli
Fail
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: HIGHPRIVILEGE_ESCALATIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PRIVILEGE_ESCALATION]: The skill contains explicit instructions for the agent to 're-run it with escalated permissions' if the sandbox blocks connection (e.g., runtime_access_denied). This encourages the agent to bypass security constraints by requesting elevated system privileges.\n- [COMMAND_EXECUTION]: The primary function of the skill is to execute shell commands using the orca CLI and its environment-specific variants (orca-ide, orca-dev) to perform file system and repository management.\n- [INDIRECT_PROMPT_INJECTION]: The skill uses the command 'ORCA skills get orca-cli' to dynamically load instructions at runtime, which allows external output to influence agent behavior.\n
- Ingestion points: Instructional content is pulled from the output of the ORCA binary in SKILL.md.\n
- Boundary markers: No delimiters are present to separate the dynamically loaded guide from the agent's existing instructions.\n
- Capability inventory: The skill enables the agent to execute shell commands and manage files through the CLI tool.\n
- Sanitization: The instructions do not define any sanitization or validation logic for the content returned by the executable.
Recommendations
- AI detected serious security threats
Audit Metadata