pikvm
Pass
Audited by Gen Agent Trust Hub on Jun 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface
- Ingestion points: The skill instructions enable the agent to interact with PiKVM APIs that retrieve untrusted external data, specifically device logs via the
/api/logendpoint (documented in references/api-system.md) and OCR-recognized text from video snapshots via the/api/streamer/snapshot?ocr=1endpoint (documented in references/api-streamer.md). - Boundary markers: There are no boundary markers or explicit instructions provided to the agent to treat this external content as untrusted or to ignore embedded instructions within the logs or OCR output.
- Capability inventory: The skill allows for significant system interaction, including building and running Ansible modules, performing network requests via
httpxandopen_url(references/ansible-module-implementation.md), and executing CLI commands for collection management (references/ansible-collection-setup.md). - Sanitization: The implementation patterns do not demonstrate any sanitization, escaping, or validation of the text returned from logs or OCR before it is processed by the agent.
Audit Metadata