The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted codebase content, pull request descriptions, and change sets, creating a potential surface for indirect prompt injection.
Ingestion points: Processes external files, diffs, and change descriptions from the user or source control.
Boundary markers: The instructions lack explicit delimiters or safety warnings to ensure the agent ignores instructions embedded within the data being analyzed.
Capability inventory: The skill uses shell commands and dependency mapping tools to process these inputs.
Sanitization: No explicit sanitization or filtering of the ingested content is defined in the workflow.
[EXTERNAL_DOWNLOADS]: The analysis workflow utilizes npx to execute the madge tool for generating dependency graphs. This involves downloading the package from the npm registry at runtime.
Evidence: Usage of npx madge in reference/analysis-techniques.md.
[COMMAND_EXECUTION]: The skill uses several local shell commands to perform its core functions of finding files and analyzing dependency structures.
Evidence: Commands such as grep, find, and npx madge are included in the instructions.

ripple