The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes dynamic context injection (the '!' command syntax) in SKILL.md to execute shell scripts automatically when the skill is loaded. These scripts check environment variables and search the file system for configuration files.
Evidence: The SKILL.md file contains a dynamic block executing grep, git, and path manipulation commands to resolve the FINTEL_API_KEY.
[DATA_EXFILTRATION]: The skill's initialization logic performs automated scanning of sensitive file paths, specifically searching for .env files in the current working directory and the git repository root.
Evidence: The shell command in SKILL.md attempts to locate and read the contents of .env files to extract credentials.
[PROMPT_INJECTION]: The skill ingests and processes data from the external Fintel API (api.fintel.io), which creates a surface for indirect prompt injection where instructions could be embedded in market data.
Evidence:
Ingestion points: Multiple curl commands in SKILL.md and references/api-reference.md pull data from https://api.fintel.io/v1/*.
Boundary markers: The instructions lack explicit delimiters or warnings for the agent to disregard potential instructions within the API response.
Capability inventory: The skill has access to shell execution (curl, grep, git).
Sanitization: No validation or sanitization of the API data is mentioned before processing.

fintel-data