The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains logic in SKILL.md that explicitly instructs the agent to read an external file (.ai-factory/skill-context/aif-build-automation/SKILL.md) and treat its contents as 'MANDATORY' overrides. It states that if a conflict occurs, 'the skill-context rule wins', creating a mechanism where instructions in project data can override the skill's core logic.
Ingestion points: SKILL.md (Step 0) reads .ai-factory/skill-context/aif-build-automation/SKILL.md from the project repository.
Boundary markers: Absent. No delimiters or warnings are used to prevent the agent from obeying instructions embedded in the project-specific context file.
Capability inventory: The skill possesses the ability to create and modify executable build scripts (Makefile, Taskfile.yml, justfile, magefile.go) and documentation files (README.md, AGENTS.md) using Write and Edit tools.
Sanitization: Absent. Content from the context file is interpreted as direct instructions without validation or escaping.
[COMMAND_EXECUTION]: The skill generates build automation files that define and execute shell commands for compilation, testing, and deployment. While these commands are based on standard templates, the skill can adapt them based on project analysis.
[EXTERNAL_DOWNLOADS]: The generated build files include targets for installing project dependencies via standard package managers such as npm, pip, composer, and Go modules, which involve fetching code from remote registries.

aif-build-automation