orca-per-workspace-env

Warn

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill directs the agent to identify and execute local binaries including orca, orca-ide, and orca-dev. It provides logic for selecting these executables based on environment variables like ORCA_CLI_COMMAND and ORCA_DEV_REPO_ROOT, or the host operating system context.
  • [PRIVILEGE_ESCALATION]: The instructions explicitly tell the agent to re-run commands with escalated permissions if a runtime_access_denied error is encountered, which may lead to unauthorized system changes if the underlying tool is compromised.
  • [DYNAMIC_EXECUTION]: The skill functions as a 'discovery stub' that retrieves its primary logic and documentation at runtime by executing ORCA skills get orca-per-workspace-env. This behavior allows the skill to execute content that is not present in the static source file.
  • [INDIRECT_PROMPT_INJECTION]: The dynamic loading of instructions from a CLI tool output creates an attack surface where external data influences the agent's control flow.
  • Ingestion points: Output from the ORCA skills get command in SKILL.md.
  • Boundary markers: None present; the instructions are loaded directly into the session context.
  • Capability inventory: Includes binary resolution, shell command execution, and privilege escalation requests.
  • Sanitization: No mechanisms are provided to validate or sanitize the content returned by the local executable before it is processed by the agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Oct 9, 2026, 02:38 AM