The Agent Skills Directory

[SAFE]: The skill is composed entirely of markdown documentation and instructional content. It contains no executable scripts, shell commands, or automated code.- [SAFE]: Implements proactive security measures against indirect prompt injection in the external-research.md reference, instructing the agent to treat all fetched content as untrusted data and explicitly forbidding the execution of instructions embedded within such data.- [SAFE]: Establishes data privacy safeguards by requiring the agent to ensure that secrets, tokens, credentials, and internal hostnames are excluded from any requests sent to external services during research activities.- [SAFE]: Enhances operational reliability by requiring the agent to verify all factual premises and uncertainties at their authoritative sources (the environment, official documentation, or the human user) rather than relying on potentially inaccurate or stale internal knowledge.

professional-behavior