← Back to Skills Marketplace
nogravedev

Agent Core

by nograve.dev · GitHub ↗ · v1.2.0 · MIT-0
cross-platform ✓ Security Clean
296
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install agent-core
Description
Complete AI agent operating system — 3-layer memory, heartbeat automation, cron patterns, safety rules, and drop-in workspace configs. Turn any agent from ch...
README (SKILL.md)

Agent Core — The AI Agent Operating System

The exact operating system running a production AI agent 24/7. Memory that persists across sessions. Heartbeats that check in without being asked. Safety rules that prevent disasters. Built from months of real production use.

What's Included (Full Kit)

  • 3-layer memory system — daily logs, topic memory, long-term curated memory with nightly consolidation
  • 6 drop-in workspace configs — AGENTS.md, SOUL.md, USER.md, IDENTITY.md, TOOLS.md, HEARTBEAT.md
  • Heartbeat automation — email monitoring, calendar alerts, RAM checks, project status, weather
  • 5 cron job templates — memory consolidation, proactive check-ins, task monitors, scheduled reports
  • 4 advanced patterns — browser hygiene, channel memory, multi-channel setup, RAM monitoring
  • Safety rules — anti-loop protection, sub-agent delegation, group chat etiquette, deploy safety
  • Quickstart guide — running in 10 minutes

Why This Exists

Most AI agents are stateless chatbots. They forget everything between sessions, never check in on their own, and have no guardrails. Agent Core fixes all of that.

This system was built over months of running a production agent on a Mac mini — handling emails, calendars, deployments, trading, and creative work 24/7. Every pattern here solved a real problem.

Results

  • Agent remembers context across unlimited sessions
  • Proactive check-ins 2-4x daily without being asked
  • Zero data loss from memory consolidation
  • No runaway loops or destructive operations
  • Works across iMessage, Discord, Telegram, WhatsApp, Slack

Get Agent Core

$39 — Full kit with all configs, templates, patterns, and quickstart guide.

👉 https://clawkits.gumroad.com/l/agent-core

$49 — ClawKits Complete Bundle (Agent Core + Trading Desk): 👉 https://clawkits.gumroad.com/l/clawkits-bundle

Author

Built by ClawKits — production-tested systems for AI agents. https://clawkits.xyz

Usage Guidance
This skill is a documentation/config template bundle and contains no code or installers — low technical risk. However: - It is high-level and vague about how it achieves integrations (email, calendars, messaging platforms). If you follow its templates to enable automatic agent actions, you'll need to supply credentials and connectors separately; review those connectors carefully. - The SKILL.md contains paid product links (Gumroad). This skill appears to be marketing/documentation rather than a turnkey automated installer. - Before granting an agent autonomy using these patterns, test them in a sandbox so you can verify the safety rules actually prevent harmful behavior. Install only if you understand it is guidance/templates (not an automated installer) and you are comfortable configuring any external accounts/connectors the patterns require.
Capability Analysis
Type: OpenClaw Skill Name: agent-core Version: 1.2.0 The provided files (_meta.json and SKILL.md) contain only metadata and marketing documentation for a third-party agent framework. There is no executable code, scripts, or functional instructions for the AI agent to perform. The content describes features like memory management and heartbeat automation but does not implement them or include any indicators of malicious intent or data exfiltration.
Capability Assessment
✓ Purpose & Capability
The skill is a documentation/config-template pack for running an agent OS (memory, heartbeats, cron patterns, safety rules). It declares no binaries, env vars, or installs, which is coherent for a purely instructional/template kit.
ℹ Instruction Scope
SKILL.md is primarily descriptive/marketing and lists features, templates, and purchase links rather than precise runtime commands. That keeps scope narrow, but the doc is high-level and vague about how integrations (email, calendars, iMessage, Slack, Telegram, etc.) are achieved — those integrations would require credentials/config not supplied by this skill.
✓ Install Mechanism
No install spec or code files are present. This instruction-only format writes nothing to disk and does not pull external code, which is low risk.
✓ Credentials
The skill requests no environment variables or credentials. While it claims support for many external services, the lack of requested secrets is consistent with an instruction/template pack (it likely expects the user or other tooling to supply service credentials).
✓ Persistence & Privilege
The skill does not request always-on status, does not include install-time setup, and is user-invocable. It does not ask to modify other skills or system configuration.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install agent-core
  3. After installation, invoke the skill by name or use /agent-core
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.2.0
Stripped to preview — full kit on Gumroad
v1.1.0
Added premium kit upsell links
v1.0.0
Initial release — 3-layer memory, heartbeat automation, cron patterns, safety rules, drop-in workspace configs
Metadata
Slug agent-core
Version 1.2.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 3
Frequently Asked Questions

What is Agent Core?

Complete AI agent operating system — 3-layer memory, heartbeat automation, cron patterns, safety rules, and drop-in workspace configs. Turn any agent from ch... It is an AI Agent Skill for Claude Code / OpenClaw, with 296 downloads so far.

How do I install Agent Core?

Run "/install agent-core" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Agent Core free?

Yes, Agent Core is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Agent Core support?

Agent Core is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Agent Core?

It is built and maintained by nograve.dev (@nogravedev); the current version is v1.2.0.

More Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463283 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259410 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200423 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191115 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190156 · by oswalpalash

💬 Comments