ota
Fail
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill instructs the agent to download and execute installation scripts from the vendor's distribution domain (dist.ota.run) using curl piped to sh and irm piped to iex.
- [EXTERNAL_DOWNLOADS]: Fetches installation assets and GitHub Actions from ota-run infrastructure.
- [COMMAND_EXECUTION]: The agent executes local system commands through the ota CLI for contract validation and task automation.
- [INDIRECT_PROMPT_INJECTION]: The skill processes repository files like ota.yaml and AGENTS.md, which may contain instructions influencing agent behavior.
- Ingestion points: ota.yaml, ota.workspace.yaml, AGENTS.md, and CLAUDE.md.
- Boundary markers: Guidelines exist to exclude prose and self-origin docs from detection evidence.
- Capability inventory: The tool provides capabilities for shell command execution and filesystem modification.
- Sanitization: Validation is performed via the Ota tool's internal contract logic.
Recommendations
- HIGH: Downloads and executes remote code from: https://dist.ota.run/install.sh - DO NOT USE without thorough review
Audit Metadata