@dodge1218

venture-spawner manages concurrent sub-agent lifecycle from job board intake through completion tracking. enforces concurrency limits, passes typed context by bucket, handles dependencies, and validates acceptance criteria before marking work done.

skill-naming-standard provides naming conventions and patterns for agent skills, covering display names, slugs, and approval boundaries. enforces clarity over jargon and requires decision validation before applying.

active-self-improvement scans logs for repeated errors, corrections, and emerging patterns, then proposes targeted edits to skills and protocols with risk classification. applies low-risk changes immediately, flags high-risk ones for approval, and feeds error context back into skill definitions.

dsb-auto-improve scans error logs and learnings to detect patterns, then applies updates to skills and memory with risk-based gate logic. lacks concrete decision criteria and rollback mechanics.

Full batch processor for corpus-scale task execution. Handles Google Drive dumps, ChatGPT exports, Apple Notes, or any large collection of mixed content (ide...

Memory health monitoring, integrity checks, and 3-layer memory architecture for AI agents. Use when agents need to prevent memory loss, detect context overfl...