The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process various project materials such as Specs, task records, and research notes, which may contain untrusted data.
Ingestion points: The agent is instructed to read relevant materials, project documentation, code, tests, and contracts (SKILL.md).
Boundary markers: There are no explicit instructions for the agent to use delimiters or to ignore potential instructions embedded within the materials it reads.
Capability inventory: The skill is authorized to perform file updates, correct references, and suggest or execute Git operations including commits and deletions (SKILL.md).
Sanitization: The instructions do not define specific sanitization or filtering steps for the content processed from external files.
[EXTERNAL_DOWNLOADS]: The references/workflow-materials.md file references external documentation from Matt Pocock's skills repository on GitHub. These links are provided for informational purposes to define upstream responsibilities and do not involve automated script execution or package installation.

knowledge-steward