The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The scripts/configure.py script facilitates the storage of the ALI_1688_AK API key in a local plaintext configuration file located at ~/.openclaw/openclaw.json. While common for developer tools, this exposes the credential to any process with read access to the user's home directory.
[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by interpolating untrusted product data from the 1688 API into the agent's context.
Ingestion points: Untrusted product titles, descriptions, and statistics are retrieved through scripts/api.py and scripts/search.py.
Boundary markers: The skill does not use delimiters or provide specific instructions to the agent to ignore potentially malicious instructions embedded in the product data.
Capability inventory: The skill has the capability to execute Python scripts, write to the filesystem, and perform network operations.
Sanitization: No validation, escaping, or filtering is performed on the data returned from the 1688 API before it is rendered to the user.
[COMMAND_EXECUTION]: The skill instructions guide the agent to manage environment variables and execute various local Python scripts (configure.py, search.py, publish.py) using the python3 binary.
[EXTERNAL_DOWNLOADS]: The scripts/api.py module makes network requests to the official 1688 platform via ainextpre.1688.com to perform product searches and synchronize listing data with downstream ecommerce stores.

1688-skills