The Agent Skills Directory

[DYNAMIC_EXECUTION]: The script scripts/__init__.py utilizes importlib.import_module to lazily load modules from computed paths. While the module names are derived from a hardcoded mapping, dynamic loading represents a technique that can be leveraged to execute code at runtime from non-static paths.\n- [CREDENTIALS_UNSAFE]: In scripts/data_service.py, the EntDataService class performs API requests using requests.get with the verify=False parameter. This disables SSL/TLS certificate validation, making the communication channel vulnerable to Man-in-the-Middle (MITM) attacks. This flaw directly risks the exposure of the QIBOOK_ACCESS_KEY sensitive environment variable to unauthorized entities.\n- [INDIRECT_PROMPT_INJECTION]: The skill possesses a significant attack surface for indirect prompt injection.\n
Ingestion points: The fetch_all_data function in scripts/data_service.py retrieves data from an external enterprise API.\n
Boundary markers: The markdown templates (e.g., templates/company.md) lack explicit boundary markers or instructions for the agent to disregard instructions potentially embedded within the data fields.\n
Capability inventory: The skill is configured with powerful tools including Bash, Read, Grep, and Glob, which increases the potential impact if the agent follows malicious instructions in the processed data.\n
Sanitization: There is no evidence of sanitization, filtering, or escaping of the JSON data retrieved from the API before it is passed to the LLM for report generation.

qibook-company-wiki-deepresearch