The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a significant surface for indirect prompt injection due to its requirement to ingest and obey instructions from external project files. Specifically, it treats rules found in .ai-factory/skill-context/aif-docs/SKILL.md as mandatory overrides that take precedence over its own logic. This creates a high-trust pathway for malicious instructions embedded in a project to manipulate agent behavior.
Ingestion points: The skill reads .ai-factory/skill-context/aif-docs/SKILL.md, README.md, docs/*.md, and project manifest files like package.json or requirements.txt.
Boundary markers: There are no specified delimiters or instructions to ignore nested directives within the ingested files.
Capability inventory: The skill can create directories, write/edit files, execute npx and python commands via Bash, and perform web searches/fetches.
Sanitization: No sanitization or validation of the ingested content is performed before it is integrated into the agent's context or used to generate new documentation artifacts.

aif-docs