aliyun-docmind-extract
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted document data through an AI parsing service, which poses a risk of indirect prompt injection if the documents contain malicious instructions.
- Ingestion points: Processes documents from URLs or local file paths via environment variables such as
DOCMIND_FILE_URLand script logic usingfs.createReadStream(SKILL.md, scripts/quickstart.js). - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the interface definition (agents/openai.yaml).
- Capability inventory: The skill environment has network access to communicate with the DocMind API and file system read capabilities (scripts/quickstart.js).
- Sanitization: Extracted text and document structures are returned directly from the API to the agent's context without evident sanitization or filtering.
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of external dependencies from the NPM registry.
- Evidence: Instructs the user to install official packages:
@alicloud/docmind-api20220711,@alicloud/tea-util, and@alicloud/credentials(SKILL.md). These are well-known packages from a recognized cloud provider.
Audit Metadata