The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes several local shell and PowerShell scripts for operational tasks:
setup-product-implement.sh and setup-product-implement.ps1: Initialize the directory structure and check for existing PDR files.
validate-pdr.sh and validate-pdr.ps1: Perform validation checks on individual PDR files.
validate-prd.sh and validate-prd.ps1: Ensure the generated PRD matches required standards (Mermaid diagrams, section numbering, traceability).
migrate-pdr-frontmatter.sh: A maintenance script that performs file system operations such as moving, deleting, and editing Markdown files to repair historical workspace inconsistencies.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data in the form of PDR files, which serves as a potential surface for indirect prompt injection.
Ingestion points: Reads all PDR-*.md files from the .adlc/drafts/pdr/ directory for analysis and content generation as described in SKILL.md (Step 1.1).
Boundary markers: Uses mandatory section templates and placeholders to structure the output, which helps mitigate accidental instruction following but does not provide complete isolation.
Capability inventory: The skill has permissions to read and write files within the repository (docs/adlc/ and .adlc/) and to execute local scripts.
Sanitization: There is no explicit sanitization or escaping of the content read from PDR files before it is interpolated into the PRD generation phase.

product-implement