The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The script scripts/xiaohongshu_analyzer.py retrieves the REDFOX_API_KEY from environment variables and transmits it in headers to https://redfox.hk. The script explicitly disables SSL certificate validation (ssl.CERT_NONE and check_hostname = False), which makes the API key vulnerable to interception via Man-in-the-Middle (MitM) attacks during transmission.
[DATA_EXFILTRATION]: The skill performs network operations, sending Xiaohongshu account identifiers to https://redfox.hk. While this is core functionality and targets the vendor's own domain, it constitutes external data transmission to a non-whitelisted domain.
[COMMAND_EXECUTION]: The skill utilizes several internal Python scripts (scripts/xiaohongshu_analyzer.py, scripts/html_generator.py, scripts/html_checker.py) for data retrieval and report generation. These are executed via the shell within the agent environment.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted data from Xiaohongshu accounts (such as nicknames and post titles) and includes them in the analysis process without robust sanitization.
Ingestion points: Data is fetched by scripts/xiaohongshu_analyzer.py from the redfox.hk API.
Boundary markers: The skill lacks explicit delimiters or instructions to ignore embedded commands within the processed account data.
Capability inventory: The skill has the ability to write files to the output/ directory, perform network requests, and execute shell commands.
Sanitization: There is no evidence of input validation or escaping for the data interpolated into the diagnostic reports or HTML templates.

xiaohongshu-account-analyzer