The Agent Skills Directory

[SAFE]: The skill consists of instructional markdown files (SKILL.md, readme-review-checklist.md, readme-template.md) that guide the agent in formatting documentation. There are no scripts, binaries, or network-enabled commands included.- [PROMPT_INJECTION]: Indirect variant. The skill is designed to process existing README files which are untrusted inputs. 1. Ingestion points: The agent reads existing README.md files from the repository during 'improvement' or 'review' tasks. 2. Boundary markers: The skill does not define explicit delimiters to separate untrusted README content from instructions. 3. Capability inventory: The skill relies on standard agent capabilities for file system access (reading and writing documentation). 4. Sanitization: No sanitization of input content is performed. This finding is considered safe as the skill lacks network or code execution capabilities.

deepresearch-readme