The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to read existing repository files (glossary and ADRs) and project-specific injected context to perform updates and de-duplication.
Ingestion points: Workflow step 2 in SKILL.md specifies reading existing files, and the 'Consumer bindings' section notes that project-specific context is appended during the skill read.
Boundary markers: There are no specific delimiters or instructions defined to prevent the agent from following malicious instructions that might be embedded in the files being read.
Capability inventory: The skill performs file-write operations to persist architectural decisions and glossary terms.
Sanitization: No sanitization or content validation logic is described for the data being processed and persisted.
[EXTERNAL_DOWNLOADS]: The skill references several external Markdown files hosted on GitHub for formatting and schema definitions.
Fetches glossary-format.md, adr-format.md, and output-schema.md from the author's toolbox repository to guide output generation.

domain-model