The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute local commands including git status, git log, git blame, rg, grep, and find. These are standard utilities for code analysis and are used within their intended functional scope for identifying changes and searching the project's lexicon.
[INDIRECT_PROMPT_INJECTION]: The skill analyzes code and documentation files (e.g., UBIQUITOUS_LANGUAGE.md, CLAUDE.md, or the project README) which are external data sources that could be controlled by an attacker. While these could potentially contain malicious instructions intended to influence the agent's behavior, the skill's operations are read-only and its output is strictly limited to identifier naming recommendations.
Ingestion points: Local repository source files, git diffs, and project documentation files accessed via rg, grep, and find.
Boundary markers: None explicitly defined in the instructions to separate the data (code/docs) from the agent's internal reasoning.
Capability inventory: Local shell execution of git, ripgrep, grep, and find for reading purposes.
Sanitization: The skill does not perform explicit sanitization or filtering of the file contents before processing them for naming analysis.

naming-review