fast-track-ai-init
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local shell scripts (
fast-track-ai-init.shandfast-track-ai-index.sh) to create directory structures and generate an index file. These scripts include robust input validation, such as regex checks for directory names and frontmatter parsing logic. - [INDIRECT_PROMPT_INJECTION]: The skill populates the
AGENTS.mdfile with a managed block of instructions that direct the AI on how to handle project files and interpret specific user inputs (e.g., using '.' as a 'proceed' command). These instructions are a core part of the skill's intended workflow design and are encapsulated within clear boundary markers. - Ingestion points:
AGENTS.md(viaassets/AGENTS-snippet.md) - Boundary markers:
<!-- BEGIN MANAGED BLOCK: fast-track-ai -->and<!-- END MANAGED BLOCK: fast-track-ai -->markers are used to delimit the instructions. - Capability inventory: The skill can create directories, write files, and perform repository indexing via the provided shell scripts.
- Sanitization: The shell scripts validate frontmatter keys and values, directory naming conventions, and file existence before indexing.
Audit Metadata