Skills
Navigate
skills/mvanhorn/printing-press-library/pp-ollama-cloud/Gen Agent Trust Hub

pp-ollama-cloud

Pass

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to download and install components using npx and go install from vendor-controlled repositories.
  • [REMOTE_CODE_EXECUTION]: Installation procedures involve executing remote code via npx for the @mvanhorn/printing-press-library package.
  • [DATA_EXFILTRATION]: The CLI provides a --deliver flag that can be used to send command results, potentially containing sensitive prompt data, to arbitrary URLs via the webhook sink.
  • [DATA_EXFILTRATION]: A feedback feature is present that can transmit local log data to a remote endpoint defined by an environment variable.
  • [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection. Ingestion points: Data enters the system via the chat and advise command prompt inputs. Boundary markers: The instructions lack delimiters or instructions to ignore embedded commands in the input data. Capability inventory: The skill can perform network POST requests and write to the local filesystem through its delivery system. Sanitization: There are no documented mechanisms for sanitizing or validating external input.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 1, 2026, 02:05 PM