req-track

Pass

Audited by Gen Agent Trust Hub on Nov 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the meegle and lark-cli command-line tools to fetch work item details, list operation records, and update Lark documents. These commands are integral to the skill's primary function of demand tracking and status reporting.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from external work items and user comments which may contain malicious instructions designed to manipulate the agent's behavior during summary generation or when writing data back to documents.
  • Ingestion points: Data is ingested from the output of meegle workitem get and meegle comment list within SKILL.md.
  • Boundary markers: The skill does not implement explicit delimiters or instructions to ignore nested prompt content when processing the fetched comments.
  • Capability inventory: The skill possesses write capabilities through meegle comment add and document modification via lark-cli docs +update.
  • Sanitization: No sanitization, filtering, or validation is performed on the content retrieved from external sources before it is interpreted or re-written to other systems.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 11, 2026, 02:12 AM