The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill includes a hardcoded API key (123456) in the curl setup instructions within SKILL.md. While it may be a placeholder, hardcoding credentials in the skill definition is an unsafe practice.
[DATA_EXFILTRATION]: The skill performs network operations to an external IP address (43.167.234.49) to fetch financial data. While this aligns with the skill's stated purpose, communication with a raw IP address that is not a well-known service or trusted organization is considered a risk for data exfiltration.
[COMMAND_EXECUTION]: The skill provides and encourages the execution of curl commands in a shell environment to retrieve stock data, which could be exploited if parameters are not properly sanitized.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it ingests untrusted data from an external API and incorporates it into the agent's context.
Ingestion points: Data is fetched from multiple endpoints on http://43.167.234.49:3101/api/v2/cnstock/ (defined in SKILL.md).
Boundary markers: Absent. The skill does not define any delimiters or instructions for the agent to ignore embedded commands within the fetched data.
Capability inventory: The skill utilizes shell-based data retrieval via curl (specified in SKILL.md).
Sanitization: Absent. There is no evidence of filtering or validation of the content returned from the external API before it is processed by the agent.

cn-fundamental