The Agent Skills Directory

[SAFE]: The skill contains no executable code or scripts, relying solely on natural language instructions to guide the agent's behavior.
[SAFE]: No unauthorized network operations, credential access, or privilege escalation attempts were detected. The file write operation is restricted to a local directory (~/.claude/execution/) intended for the agent's output.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from a 'decision record' to generate plans. While this is a data ingestion surface, the risk is assessed as safe because the skill's capabilities are limited to generating structured markdown and writing it to a specific local path, without any high-risk capabilities like network access or shell execution.

execute