lark-push

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes lark-cli to transmit messages and python3 to generate interactive card payloads. It also provides an optional Git hook script that monitors local commits to automate notifications.
  • [EXTERNAL_DOWNLOADS]: The documentation provides instructions for installing the official @larksuite/cli from the npm registry and Python 3 via standard system package managers.
  • [DATA_EXFILTRATION]: The skill's primary function is to collect project metadata—such as Git commit hashes, branch names, and author information—and transmit it along with user-provided report content to a configured Feishu/Lark chat group.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Git logs, standard input, and local files. It mitigates injection risks by using Bash arrays for command construction and JSON serialization for card payloads, ensuring that external data is treated as content rather than executable instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 02:19 AM
Security Audit — agent-trust-hub — lark-push