The Agent Skills Directory

[CREDENTIALS_UNSAFE] (HIGH): A hardcoded API key ctx7sk-d6069954-149e-4a74-ae8f-85092cbfcd6f is used as a default value in scripts/context7.py. Hardcoding credentials allows anyone with access to the code to use the service on the author's behalf and is a major security risk.
[DATA_EXFILTRATION] (LOW): The script communicates with https://context7.com/api/v2. While this is consistent with the skill's stated purpose of fetching documentation, it involves sending search queries and identifiers to an external, non-whitelisted domain.
[COMMAND_EXECUTION] (LOW): The skill relies on executing a local Python script (scripts/context7.py) to perform its tasks. This is the intended design but represents a local command execution surface.
[PROMPT_INJECTION] (LOW): (Category 8: Indirect Prompt Injection) The skill fetches and processes library documentation from an external API, which could be manipulated by an attacker to include malicious instructions.
Ingestion points: The get_context function in scripts/context7.py returns documentation text from the Context7 API.
Boundary markers: Absent. There are no delimiters or instructions provided to the agent to treat the fetched documentation as untrusted data.
Capability inventory: The agent can execute local Python scripts and make network requests through this skill.
Sanitization: None. The API response is returned as-is to the agent's context.

context7-2