The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown-based instructions, assessment tables, and reporting templates. It does not include or reference any scripts (Python, Node.js, Shell), binaries, or configuration files that would execute on a host system.\n- [NO_CODE]: The skill's functionality is limited to natural language processing and structured data formatting. It does not request any dangerous capabilities such as file system access, network communication, or environment variable reading.\n- [INDIRECT_PROMPT_INJECTION]: The skill defines several inputs for untrusted data (e.g., 'Change description', 'Stakeholder list'). While these ingestion points exist, the skill lacks any 'active' tools (exec, network, file-write) that could be exploited via injection. The risk is limited to the agent outputting malicious text, which is mitigated by the platform's standard safety guardrails. The severity is internally assessed as SAFE/LOW.

change-management