The Agent Skills Directory

[SAFE]: The skill consists of text-based templates and instructional guidelines for writing blameless postmortems. No evidence of command execution, credential access, or malicious persistence was found.
[INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes external data. Ingestion points: Incident descriptions, logs, and timelines provided by the user in the agent context. Boundary markers: None explicitly defined in the instructions for separating untrusted user data from the template logic. Capability inventory: The skill does not utilize any subprocess calls, file-writing tools, or network operations. Sanitization: There are no instructions for sanitizing or escaping embedded instructions within the processed incident data.

postmortem-writer