The Agent Skills Directory

[SAFE]: The skill's logic is transparent and focuses on its primary purpose of lead generation and TAM management without any detected malicious patterns.
[COMMAND_EXECUTION]: The skill provides utility scripts (tam_builder.py, backfill_company_type.py, recover_sixtyfour_people.py) meant for CLI use. These scripts follow best practices, such as providing --dry-run and --sample modes to allow users to preview actions before committing changes.
[DATA_EXPOSURE]: The skill correctly handles sensitive configuration by instructing users to store API keys and database credentials in a .env file. It accesses these variables at runtime for legitimate communication with Apollo and Supabase services.
[CREDENTIALS_UNSAFE]: No hardcoded secrets were found in the codebase. Documentation uses standard placeholders (e.g., your-api-key-here) for environment setup instructions.
[PROMPT_INJECTION]: The instructions in SKILL.md are focused on operational guidance and do not contain attempts to bypass AI safety filters or override system instructions.
[INDIRECT_PROMPT_INJECTION]: While the skill ingests data from external APIs (Apollo, SixtyFour), it processes this data as structured JSON objects mapped to specific database columns, posing no threat of executing injected instructions at runtime.

tam-builder