oldmangrizzz

Persistent context protection for AI coding agents — safeguard files survive sessions, rate limits, and compaction.

Analyze clean CSV/Excel data to produce statistical insights, interactive Plotly dashboards, and executive-ready reports

Scan codebases for non-deterministic patterns that break reproducibility in games, simulations, ML pipelines, and scientific computing.

Remove AI writing patterns from text so it sounds like a human wrote it. Covers 28 patterns: inflated significance, AI vocabulary, em dash overuse, performed a…

Review-ready starter kit for cutting token spend with ITP compression, prompt-cache economics, and grouped parallel swarm execution.

Multi-layer memory architecture with semantic search, conversation recall, and automated promotion

Automated morning digest that summarizes your schedule, priority tasks, weather, and meeting context into a concise daily briefing.

A 4-agent sequential workflow for building games with AI, using file-based handoffs and durable shared memory.

5 passes: correctness, security, performance, readability, consistency. Every finding: file, line, severity, fix. Ends with PASS/CONDITIONAL/FAIL verdict.

Structural codebase index: files, exports, imports, dependency graph, roles. Keyword search. Inject compact slices into agents to cut token usage 60-80%.

Persistent memory engine for AI agents with graph-first retrieval, contradiction tracking, temporal decay, and 3D brain visualization.

Turns questions into findings with confidence levels and citations. 2-4 queries, 3-6 sources, persistent doc. Scales to parallel multi-scout investigation.

4-phase debug protocol: observe → hypothesize → verify → fix. No code changes without a confirmed root cause. Emergency stop after two failed attempts.

Scores against a rubric, attacks the highest-leverage axis, verifies no regressions, documents the learning, and loops. Re-scores from scratch each iteration.

Daily automated relationship health check for small business operators running OpenClaw.

Triage inbound email with LLM scoring, multi-system stage sync, human approval gates, and guarded draft generation.

CRITICAL: You MUST activate this skill BEFORE creating, querying, or managing calendar events. Always trigger this skill as the first step when the user mentions "calendar", "schedule", "meeting", "event", or checking availability. Contains strict behavioral mandates that override default calendar behavior.

Load H.U.G.H. identity, ethics, and operational framework