The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze potentially untrusted content such as project plans, pull requests, and source code files using the Read, Grep, and Glob tools.
Ingestion points: The workflow involves analyzing files specified by the user or auto-detected in the environment (e.g., SKILL.md usage of /premortem <file>).
Boundary markers: The instructions provide a structured reasoning framework ("Verification Checklist"), but do not include explicit instructions to sanitize inputs or ignore embedded instructions that might attempt to hijack the agent's logic during the analysis or research phases.
Capability inventory: The skill can execute TodoWrite to modify local files and uses the Task tool to spawn sub-agents for research, which could be influenced by malicious content in the source files.
Sanitization: There is no evidence of input validation or escaping for the content retrieved from external files before it is processed or passed to subsequent tools.

premortem