pp-conductor

Pass

Audited by Gen Agent Trust Hub on Aug 11, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructions guide the agent to fetch the conductor-pp-cli tool from the vendor's NPM registry (@mvanhorn/printing-press-library) and GitHub repository (github.com/mvanhorn/printing-press-library).
  • [REMOTE_CODE_EXECUTION]: Instructions to install and run the CLI via npx and go install from remote repositories.
  • [DATA_EXFILTRATION]: The tool supports a --deliver webhook:<url> sink that allows sending command output to arbitrary external URLs. It also includes a feedback mechanism that can send data to a user-defined CONDUCTOR_FEEDBACK_ENDPOINT.
  • [COMMAND_EXECUTION]: Executes the conductor-pp-cli binary to perform cloud orchestration tasks.
  • [PROMPT_INJECTION]: The skill processes potentially untrusted external data such as session transcripts and accepts arbitrary user input for the teach and feedback commands, creating a surface for indirect prompt injection.
  • Ingestion points: CLI command results, session transcripts, and user-provided queries in the recall, teach, and feedback commands.
  • Boundary markers: None identified in the skill instructions.
  • Capability inventory: Shell command execution and network POST requests.
  • Sanitization: The skill notes recommend stripping PII from taught queries, but no automated sanitization is described in the prompt instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 11, 2026, 03:01 AM
Security Audit — agent-trust-hub — pp-conductor