The Agent Skills Directory

[COMMAND_EXECUTION]: Employs the Bash tool to run git log, git ls-files, and stat commands to compare file modification timestamps and module contents for indexing accuracy.
[COMMAND_EXECUTION]: Reads the local application log file at ~/.claude/logs/index-sync.log to display a summary of recent synchronization activity.
[PROMPT_INJECTION]: Potential indirect prompt injection surface identified. 1. Ingestion points: Extracts metadata and counts from .claude/index.md and reads content from module README.md files. 2. Boundary markers: No delimiters or explicit warnings are used to distinguish ingested file content from instructions. 3. Capability inventory: Uses Bash and Read tools for filesystem inspection and file access. 4. Sanitization: Extracted data from the index and README files is not sanitized before being incorporated into the agent's context, though the disable-model-invocation setting provides a partial layer of mitigation for the immediate execution step.

index-status