The Agent Skills Directory

[SAFE]: The skill does not contain any malicious code, obfuscation, or persistence mechanisms. Its operations are consistent with its stated purpose of business research.
[COMMAND_EXECUTION]: The skill executes a local Python helper script (scripts/account_list_check.py) to perform data validation and coverage checks. This script uses only Python standard libraries, performs no network operations, and is limited to processing the research results generated during the session.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process potentially untrusted data from web pages and user-provided files. It includes specific defensive instructions (Step 8) to treat all fetched or pasted content strictly as data and to explicitly ignore and report any instructions embedded within that data.
[DATA_EXPOSURE]: The skill correctly manages sensitive data by instructing the agent to never guess contact details and to use only the user's existing systems for private information. It enforces a strict policy of using public professional statements for person-specific research.

account-research