The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts (analyze_access.py, track_kb_access.py) and delegates tasks to other internal skills like /kb-mint. These operations are used for log analysis and skill management within the project context.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes project documentation and access logs to generate recommendations.
Ingestion points: The analyzer agent reads index.md files from the knowledge base to classify content. analyze_access.py reads a local JSONL log file.
Boundary markers: The skill does not use explicit delimiters or warnings when the analyzer agent reads KB content.
Capability inventory: The analyzer agent has Read and Bash tools. The main skill can execute local scripts and interact with project memory.
Sanitization: The scripts use standard JSON parsing for structured data and do not execute content derived from the logs or documentation.

kb-monitor