The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute git diff commands. This is used to identify changed files and retrieve diff content for architectural review, which is a core part of its intended functionality.
[INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it ingests and processes external data (source code and ADR files).
Ingestion points: Files retrieved via git diff and ADR documents read from the docs/adr/ directory via the Read tool.
Boundary markers: No specific delimiters or safety instructions are defined for separating ingested file content from the agent's internal instructions.
Capability inventory: The agent has access to Bash (shell execution), Read (file system access), Grep, Glob, and vendor-specific MCP tools (mcp__plugin_ruflo-core_ruflo__agentdb_causal-query, mcp__plugin_ruflo-core_ruflo__agentdb_hierarchical-query, and mcp__plugin_ruflo-core_ruflo__memory_search).
Sanitization: The instructions do not specify any sanitization or validation of the content read from external files.
[SAFE]: The identified command execution and data ingestion are standard for the skill's purpose of code review and architectural compliance. All identified external resources and plugins are associated with the vendor 'ruvnet' and align with the skill's description.

adr-review