The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes several local JavaScript scripts via the node runtime, including doctor.mjs, instructions.mjs, source-quality.mjs, and create-artifact.mjs. These tools are located in the .specforge/core/scripts/ directory and represent the intended automated functionality of the SpecForge ecosystem.
[PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks as its core purpose involves ingesting and analyzing untrusted external data, such as user requests and third-party technical documentation.
Ingestion points: Processes content from 00-intake/original-request.md, external official documentation, source code, and release notes during the research stage.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore potential commands embedded within the analyzed documents.
Capability inventory: The skill possesses the ability to execute local scripts and write files to the 01-spec/ and 00-intake/ directories.
Sanitization: There are no explicit requirements for sanitizing or escaping the content retrieved from external sources before it is processed.

sf-discovery