The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands including find, grep, cat, and git log to build a profile of the project's size, age, and contributors. These diagnostic commands are scoped to project metadata and do not target sensitive system files or credentials.
[PROMPT_INJECTION]: The skill ingests data from codebase files like package.json, CLAUDE.md, and documentation, creating a surface for indirect prompt injection. 1. Ingestion points: Local project configuration and specification files read during Phase 1. 2. Boundary markers: Absent; there are no instructions for the agent to distinguish between project data and potentially malicious instructions. 3. Capability inventory: The skill is restricted to file reading and searching tools (Read, Grep, Glob) and does not have access to network-writing or administrative shell capabilities. 4. Sanitization: None; the skill processes raw file contents as they are found in the project.

methodology-advisor