The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the which command to check for the presence of the socraticode executable on the system path.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by ingesting and processing data from the local codebase.
Ingestion points: Files identified through git status or direct user input as described in the preflight algorithm in SKILL.md.
Boundary markers: None identified; the skill does not use specific delimiters or instructions to disregard embedded commands in the files it reads.
Capability inventory: The skill invokes filesystem tools (Grep, Read), shell commands (which), and multiple MCP tools (query_graph_tool, get_impact_radius_tool, semantic_search_nodes_tool, mcp__code-review-graph__build_or_update_graph_tool) as listed in the preflight algorithm and failure modes in SKILL.md.
Sanitization: No evidence of content sanitization, filtering, or validation for the code data being processed is provided.

socraticode-preflight