The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides an attack surface by ingesting and processing raw user-supplied text to drive workflow decisions and repository analysis tools.
Ingestion points: The feature_description input field described in SKILL.md.
Boundary markers: None identified; user input is processed for keyword triggers and uncertainty signals without explicit delimiters.
Capability inventory: The skill is capable of executing shell commands via ast-grep and invoking other internal skills such as majestic-devops:devops-plan and ux-brief.
Sanitization: No explicit input sanitization or validation routines are defined for the feature_description input.
[COMMAND_EXECUTION]: The skill utilizes a bash shell to run ast-grep for structural analysis of the codebase. This is a legitimate development task consistent with the skill's purpose of repository research and feature discovery.

blueprint-discovery