The Agent Skills Directory

[DYNAMIC_EXECUTION]: The documentation for the Change Log engine in references/document-change-log.md provides a shell command that uses node -e to dynamically load a JavaScript module from a computed relative path (./.agents/skills/{develop|finalise}/references/change-log.js). This pattern of loading code from variable paths at runtime is a medium-risk practice.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to validate and audit documentation files, such as PRDs, epics, and stories. This creates an attack surface where malicious instructions could be embedded in the data being processed.
Ingestion points: Documentation files (PRDs, epics, stories) read during validation and auditing tasks as described in SKILL.md.
Boundary markers: Absent; the skill does not instruct the agent to use specific delimiters or to ignore embedded instructions within the documents being validated.
Capability inventory: The skill performs file system read and write operations (via the change log engine) to update documentation and logs.
Sanitization: Absent; the skill relies on pattern matching for validation and does not appear to sanitize the content of the markdown files it processes.

documentation-standards-validator