The Agent Skills Directory

[SAFE]: The skill is purely instructional and follows safe practices for defining agent behavior and persona. No malicious patterns or security risks were identified.
[NO_CODE]: No scripts or executable files are included in this skill, eliminating risks associated with remote code execution, privilege escalation, or persistence.
[PROMPT_INJECTION]: The instructions provide guidance on writing style and brevity. There are no attempts to bypass safety filters, ignore prior instructions, or extract system prompts.
[SAFE]: Indirect prompt injection analysis:
Ingestion points: Slack thread transcripts (processed by the 'Digest' task).
Boundary markers: Not explicitly defined in the provided instructions.
Capability inventory: No executable scripts, network operations, or file system access identified in the skill files.
Sanitization: No input validation or escaping mechanisms are mentioned.
Analysis: While the skill processes untrusted external data, the risk is negligible because it lacks any dangerous capabilities that could be triggered by injection.

slack