observe-cli
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted data retrieved from external sources such as telemetry datasets, platform documentation, and other skills stored in Observe. This creates a surface for indirect prompt injection where malicious content in the data could influence the agent's behavior.
- Ingestion points: Data returned by
observe query,observe dataset view,observe skill view, andobserve docs searchcommands. - Capability inventory: The skill has the ability to execute shell commands via the
observeCLI. - Boundary markers: No delimiters or boundary instructions are provided to help the agent distinguish between data and instructions.
- Sanitization: There is no evidence of sanitization or validation of the content pulled from the platform.
- [COMMAND_EXECUTION]: The skill relies on executing shell commands through the
observeCLI to perform its tasks. It specifically includes a mandatory instruction for the agent to runobserve skill view observe-cli --contentto dynamically fetch and load its own instruction set from the platform, which is a form of runtime instruction modification.
Audit Metadata