opencode-v2-api
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines an interaction model where the agent processes external data from an API, creating a vulnerability surface for indirect instructions.
- Ingestion points: Data is ingested from the SSE event stream (
/api/event), general API responses, and file contents retrieved through/api/fs/read/*. - Boundary markers: There are no instructions provided for the agent to use delimiters or ignore potential instructions embedded within the API responses.
- Capability inventory: The API grants extensive capabilities including file system traversal/reading (
/api/fs/*), terminal session management (/api/pty/*), and the ability to send commands to active sessions (/api/session/{id}/command). - Sanitization: The documentation does not specify any sanitization or validation requirements for data returned by the server.
- [COMMAND_EXECUTION]: The documentation includes routes for executing shell commands and managing interactive PTY sessions (terminal access) on the host server via the
/api/ptyand/api/session/{id}/commandendpoints. - [NO_CODE]: The skill consists entirely of markdown documentation and YAML frontmatter; it does not include any executable scripts, binaries, or automated environment modifications.
Audit Metadata