The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes content fetched from external Alibaba Cloud domains to generate its review report, which introduces a potential surface for indirect prompt injection.
Ingestion points: The fetch_url function in scripts/review_product_docs_and_api.py retrieves data from external URLs including api.aliyun.com and www.aliyun.com.
Boundary markers: The generated markdown and JSON reports do not include explicit delimiters or instructions for the agent to ignore potentially malicious embedded content within the fetched data.
Capability inventory: The skill has the capability to write files to the local file system (output/aliyun-platform-docs-review/) via the pathlib.Path.write_text method used in the execution script.
Sanitization: The script performs structural extraction (links, JSON fields) but does not sanitize the text content retrieved from external sources before including it in the generated report.
[COMMAND_EXECUTION]: The skill executes a local Python script review_product_docs_and_api.py to perform the documentation audit. The script accepts a user-supplied product name as a command-line argument.
[EXTERNAL_DOWNLOADS]: The skill fetches product metadata and API documentation from official Alibaba Cloud domains, such as api.aliyun.com and www.aliyun.com. These are recognized as official domains for a well-known service provider.

aliyun-platform-docs-review