The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes data from user-provided arguments and existing incident reports in the '.chalk/docs/engineering/' directory, creating a surface for indirect prompt injection. Malicious instructions embedded in these reports could potentially influence the agent's behavior during the postmortem generation process. * Ingestion points: $ARGUMENTS and incident report files. * Boundary markers: Absent; no specific delimiters or instructions are used to distinguish untrusted content from the agent's primary instructions. * Capability inventory: The agent utilizes Read, Glob, Grep, and Write tools to manage engineering documentation. * Sanitization: The skill implements role-based name redaction to maintain blamelessness but does not filter for potential prompt injection patterns or instructions within the input data.

create-postmortem