The Agent Skills Directory

[COMMAND_EXECUTION]: Employs bash commands such as find, awk, sed, and wc to perform filesystem discovery, count lines, and evaluate project complexity for indexing purposes. These operations are local and consistent with the skill's stated goal of codebase documentation.
[PROMPT_INJECTION]: Susceptible to Indirect Prompt Injection (Category 8) due to its processing of untrusted project data. 1. Ingestion points: The skill reads existing CLAUDE.md files and utilizes subagents to scan source code for specific behavioral comments and project-specific rules in Phase 1 and Phase 3. 2. Boundary markers: The prompts lack explicit delimiters or instructions to treat codebase content as untrusted, which may allow malicious instructions embedded in files to influence the agent's output. 3. Capability inventory: The skill has the capability to execute shell commands, read and write files within the project directory, and orchestrate multiple subagents. 4. Sanitization: No sanitization or validation of the extracted code content is performed before it is summarized or used to guide documentation generation.

index-knowledge