pp-cdc-pakistan

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the cdc-pakistan-pp-cli and cdc-pakistan-pp-mcp tools.
  • Fetches the CLI package using npx -y @mvanhorn/printing-press-library install cdc-pakistan --cli-only from the NPM registry.
  • Fetches and compiles the Go binary using go install github.com/mvanhorn/printing-press-library/library/payments/cdc-pakistan/cmd/cdc-pakistan-pp-cli@latest from GitHub.
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to install, verify, and run the specialized CLI tool.
  • Commands include installation (npx, go install), verification (cdc-pakistan-pp-cli --version, cdc-pakistan-pp-cli doctor), and operational commands (coverage map, identity ledger, stats history, etc.).
  • [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection by processing external data and user input.
  • Ingestion points: Reads and processes PDF files and HTML tables fetched from the CDC Pakistan website via the CLI tools. It also processes natural language queries in the which, recall, and teach commands.
  • Boundary markers: The skill does not explicitly define boundary markers for the external financial data it processes.
  • Capability inventory: The skill has Read Bash tool access and can execute the cdc-pakistan-pp-cli binary, which supports network operations (webhooks) and file writing.
  • Sanitization: The CLI implements data integrity checks (verify rows, verify schema) and provides instructions for stripping PII (names, emails, account IDs) before using the teach command to update the local learning store.
  • [DATA_EXFILTRATION]: The CLI tool includes a --deliver webhook:<url> flag that allows routing command output to a user-specified HTTP endpoint. While a functional feature for data integration, it represents a path for routing internal command results to external servers.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 10:44 AM
Security Audit — agent-trust-hub — pp-cdc-pakistan