The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill reads and processes multiple untrusted project artifacts to pre-fill communication templates, creating a surface for data-driven instruction injection.
Ingestion points: The skill reads CLAUDE.local.md, .sync-watch.json, PRD, Meeting-Notes, Stakeholder Directory, Presentations, Change-Log, DR-Plan, and IRP files.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands (e.g., XML tags or clear markdown separators) are defined to protect the agent context from malicious content within those files.
Capability inventory: The skill has the capability to read local filesystem artifacts and write new markdown files to the Comms/ directory.
Sanitization: The instructions do not define any sanitization, filtering, or validation steps for the content extracted from external project files before it is interpolated into the final output.

pmstudio-comms