skills/larksuite/cli/lark-vc/Gen Agent Trust Hub

lark-vc

Pass

Audited by Gen Agent Trust Hub on Mar 28, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill facilitates integration with Lark's Video Conferencing (VC) services. It uses the lark-cli tool to interact with official Lark APIs (e.g., /open-apis/vc/v1/meetings/search).
  • [COMMAND_EXECUTION]: The skill uses lark-cli for various operations, which is the intended mechanism for this vendor-provided tool. The commands are structured and restricted to the defined resources (vc, drive, docs).
  • [EXTERNAL_DOWNLOADS]: The skill downloads meeting transcriptions and AI artifacts directly from Lark's infrastructure to a local output directory. These are part of the primary functionality (retrieving meeting records) and are directed to the user's local environment.
  • [DATA_EXFILTRATION]: While the skill reads sensitive meeting data (notes, transcripts, summaries), it does so within the authenticated context of the Lark CLI. There are no patterns suggesting the exfiltration of this data to unauthorized third-party domains. All documented endpoints belong to the larksuite ecosystem.
  • [CREDENTIALS_UNSAFE]: No hardcoded credentials were found. The skill explicitly delegates authentication to a shared module (lark-shared/SKILL.md) and requires the user to run lark-cli auth login.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 28, 2026, 02:44 AM