The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a local utility, scripts/verify.sh, used to validate the structural integrity of generated CSV calendar files. The script uses standard system tools like awk and sed in a read-only manner and does not perform any network operations or unauthorized file modifications.
[PROMPT_INJECTION]: The instructions incorporate safety guardrails via 'STOP' gates. These instructions mandate that the agent ground its planning in existing project documentation (within the 02-DOCS/ directory) and prevent it from inventing brand strategy or tone, which aligns with security best practices for grounding.
[DATA_EXFILTRATION]: No network access patterns, hardcoded credentials, or exfiltration vectors were detected. The skill operates entirely within the local project context.

content-engine