wecomcli-doc
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script,
scripts/build_docx.py, to process JSONL specifications and generate Word documents. The script implements strict input validation and path sandboxing to mitigate risks, including checking for symbolic links and restricting access to allowed directories. - [INDIRECT_PROMPT_INJECTION]: The skill reads content from external WeCom documents via the
wecom-cli doc contents getcommand, which introduces a potential surface for indirect prompt injection. - Ingestion points: Document content read in
SKILL.md. - Boundary markers: Absent; content is ingested into the context without specific delimiters.
- Capability inventory: File system writes via
build_docx.pyand document modification viawecom-cli. - Sanitization: No specific sanitization or escaping of the document content is performed before processing.
Audit Metadata