Proactive Agent
Wire a proactive, self-improving agent architecture with memory survival, recovery after compaction, and security guardrails across long sessions.
Overview
proactive-agent is a journey-wide agent skill that defines proactive, persistent, self-improving agent architecture—with WAL, buffer, compaction recovery, and security patterns—usable whenever a solo builder needs reliab
Install
npx skills add https://github.com/halthelobster/proactive-agent --skill proactive-agentWhat is this skill?
- WAL Protocol for corrections and decisions that must survive context loss
- Working Buffer and Compaction Recovery for the flush-to-compaction danger zone
- Unified Search across sources before defaulting to “I don’t know”
- Security hardening: skill install vetting, network warnings, context leakage prevention
- Relentless resourcefulness and Self-Improvement Guardrails (ADL/VFM) for safe evolution
- Version 3.0.0 with WAL Protocol, Working Buffer, and Compaction Recovery
- Three Pillars: Proactive, Persistent, and self-improving architecture
Adoption & trust: 14.4k installs on skills.sh; 17 GitHub stars; 3/3 security scanners passed (skills.sh audits).
What problem does it solve?
Your agent forgets decisions after compaction, waits for prompts, and lacks safe patterns for memory, search, and skill installation.
Who is it for?
Solo builders running long Claude Code or Cursor sessions who want institutional memory and proactive check-ins without building custom infra from scratch.
Skip if: One-shot codegen with no ongoing agent persona, or teams that forbid persistent logs and self-modifying agent behavior policies.
When should I use this skill?
Transform agents from task-followers into proactive partners; implement WAL, buffers, compaction recovery, unified search, or Hal Stack security/self-improvement patterns.
What do I get? / Deliverables
You adopt v3.0 pillars and protocols so the agent anticipates work, survives context loss, recovers after truncation, and evolves with explicit security guardrails.
- Agent architecture aligned to Three Pillars and v3.0 protocols
- Documented recovery and search behaviors after context loss
- Security and self-improvement guardrail checklist for skill use
Recommended Skills
Journey fit
Useful at every journey phase - explore requirements and options before committing to a direction.
Where it fits
Install WAL and Working Buffer conventions before a multi-day feature branch driven entirely by the agent.
Apply skill-install vetting and network warnings before pulling community skills into production repos.
Run compaction recovery when context truncation drops corrections and the agent starts contradicting prior decisions.
Use proactive check-ins and unified search so the agent monitors recurring user issues without waiting for you to paste logs.
Use reverse prompting to surface experiments you had not asked for while exploring product directions.
How it compares
Agent operating-system methodology—not a single MCP tool or a narrow debugging checklist.
Common Questions / FAQ
Who is proactive-agent for?
Indie developers and small teams treating the coding agent as a daily partner who must remember decisions, survive compactions, and surface ideas proactively.
When should I use proactive-agent?
During build when wiring agent-tooling; at ship when hardening security around skills; at operate when monitoring and recovery matter; and anytime you redesign agent memory or proactive check-ins.
Is proactive-agent safe to install?
It emphasizes vetting skill installs and leakage prevention—still review the Security Audits panel on this page and align WAL/buffer storage with your secrets policy.
SKILL.md
READMESKILL.md - Proactive Agent
# Proactive Agent 🦞 **By Hal Labs** — Part of the Hal Stack **A proactive, self-improving architecture for your AI agent.** Most agents just wait. This one anticipates your needs — and gets better at it over time. ## What's New in v3.0.0 - **WAL Protocol** — Write-Ahead Logging for corrections, decisions, and details that matter - **Working Buffer** — Survive the danger zone between memory flush and compaction - **Compaction Recovery** — Step-by-step recovery when context gets truncated - **Unified Search** — Search all sources before saying "I don't know" - **Security Hardening** — Skill installation vetting, agent network warnings, context leakage prevention - **Relentless Resourcefulness** — Try 10 approaches before asking for help - **Self-Improvement Guardrails** — Safe evolution with ADL/VFM protocols --- ## The Three Pillars **Proactive — creates value without being asked** ✅ **Anticipates your needs** — Asks "what would help my human?" instead of waiting ✅ **Reverse prompting** — Surfaces ideas you didn't know to ask for ✅ **Proactive check-ins** — Monitors what matters and reaches out when needed **Persistent — survives context loss** ✅ **WAL Protocol** — Writes critical details BEFORE responding ✅ **Working Buffer** — Captures every exchange in the danger zone ✅ **Compaction Recovery** — Knows exactly how to recover after context loss **Self-improving — gets better at serving you** ✅ **Self-healing** — Fixes its own issues so it can focus on yours ✅ **Relentless resourcefulness** — Tries 10 approaches before giving up ✅ **Safe evolution** — Guardrails prevent drift and complexity creep --- ## Contents 1. [Quick Start](#quick-start) 2. [Core Philosophy](#core-philosophy) 3. [Architecture Overview](#architecture-overview) 4. [Memory Architecture](#memory-architecture) 5. [The WAL Protocol](#the-wal-protocol) ⭐ NEW 6. [Working Buffer Protocol](#working-buffer-protocol) ⭐ NEW 7. [Compaction Recovery](#compaction-recovery) ⭐ NEW 8. [Security Hardening](#security-hardening) (expanded) 9. [Relentless Resourcefulness](#relentless-resourcefulness) ⭐ NEW 10. [Self-Improvement Guardrails](#self-improvement-guardrails) ⭐ NEW 11. [The Six Pillars](#the-six-pillars) 12. [Heartbeat System](#heartbeat-system) 13. [Reverse Prompting](#reverse-prompting) 14. [Growth Loops](#growth-loops) --- ## Quick Start 1. Copy assets to your workspace: `cp assets/*.md ./` 2. Your agent detects `ONBOARDING.md` and offers to get to know you 3. Answer questions (all at once, or drip over time) 4. Agent auto-populates USER.md and SOUL.md from your answers 5. Run security audit: `./scripts/security-audit.sh` --- ## Core Philosophy **The mindset shift:** Don't ask "what should I do?" Ask "what would genuinely delight my human that they haven't thought to ask for?" Most agents wait. Proactive agents: - Anticipate needs before they're expressed - Build things their human didn't know they wanted - Create leverage and momentum without being asked - Think like an owner, not an employee --- ## Architecture Overview ``` workspace/ ├── ONBOARDING.md # First-run setup (tracks progress) ├── AGENTS.md # Operating rules, learned lessons, workflows ├── SOUL.md # Identity, principles, boundaries ├── USER.md # Human's context, goals, preferences ├── MEMORY.md # Curated long-term memory ├── SESSION-STATE.md # ⭐ Active working memory (WAL target) ├── HEARTBEAT.md # Periodic self-improvement checklist ├── TOOLS.md # Tool configurations, gotchas, credentials └── memory/ ├── YYYY-MM-DD.md # Daily raw capture └── working-buffer.md # ⭐ Dan