pp-everbee

Pass

Audited by Gen Agent Trust Hub on Aug 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill documentation instructs the agent to install external tools from the author's infrastructure. It fetches the '@mvanhorn/printing-press-library' package from NPM via 'npx' and downloads/compiles the 'everbee-pp-cli' and 'everbee-pp-mcp' tools from 'github.com/mvanhorn/printing-press-library' using 'go install'.
  • [COMMAND_EXECUTION]: Shell commands are required for setup and operation. The skill uses 'npx -y' during installation, which bypasses package execution confirmations, and invokes the CLI binary with various flags that modify local application state.
  • [DATA_EXFILTRATION]: The CLI includes documented features for egressing data to external endpoints. This includes a 'feedback' command that can send logs to a configured 'EVERBEE_FEEDBACK_ENDPOINT' and a '--deliver webhook' sink that POSTs command outputs to a user-specified URL.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection via the processing of external market research data. The tool ingests data from the Etsy marketplace through the EverBee API and uses a 'learning loop' (recall/teach) to persist choreographies and findings in a local database. While the skill advises PII stripping and uses JSON envelopes for tool outputs, the persistent storage of externally-derived data allows it to influence future agent reasoning and command execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 22, 2026, 03:13 AM