The Agent Skills Directory

[COMMAND_EXECUTION]: The skill extensively uses the Bash tool to perform repository discovery and pattern analysis. It executes commands such as git remote get-url, find, ls, and grep to identify project structure, framework identities, and key code patterns. These operations are intended for local repository analysis.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it ingests and processes untrusted source code from the repository to generate architectural documentation.
Ingestion points: The skill uses Read and Grep tools to analyze source files, entry points, and domain patterns within the repository.
Boundary markers: There are no explicit instructions or delimiters provided to the agent to help it distinguish between legitimate code logic and potentially malicious natural language instructions embedded in code comments or strings within the analyzed repository.
Capability inventory: The skill has Write access to create or update documentation files (such as ARCHITECTURE.md and QUALITY_SCORE.md) and Bash access for repository interaction.
Sanitization: No sanitization, filtering, or validation logic is specified for the content extracted from the source files before it is interpolated into the documentation templates.

setup-harness