OpenCortex
Self-improving memory architecture for OpenClaw agents. Structured memory files, nightly distillation, weekly synthesis, enforced principles (P0 for custom, P1-P8 managed), write-ahead durability, and model-agnostic delegation — so your agent compounds knowledge instead of forgetting it. Includes opt-in metrics tracking with growth charts and compound scoring to measure effectiveness over time. All sensitive features (voice profiling, infrastructure auto-collection, git push) are OFF by default and require explicit opt-in via environment variable or flag. Safe to install: no network calls during setup, fully auditable bash scripts, isolated cron sessions scoped to workspace only. Use when: (1) setting up a new OpenClaw instance, (2) user asks to improve/organize memory, (3) user wants the agent to stop forgetting things, (4) bootstrapping a fresh agent with best practices. NOT for: runtime memory_search queries (use built-in memory tools). Triggers: "set up memory", "organize yourself", "stop forgetting", "memory architecture", "self-improving", "cortex", "bootstrap memory", "memory optimization".
Why use this skill?
Upgrade OpenClaw with OpenCortex, a self-improving memory architecture that compounds knowledge, automates distillation, and enforces operational principles.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/jd2005l/opencortexWhat This Skill Does
OpenCortex transforms your standard OpenClaw agent from a transient assistant into a persistent, self-evolving intelligence. It implements a sophisticated memory architecture designed to combat the common problem of agent amnesia. Instead of dumping all data into a single, unorganized file, OpenCortex structures your workspace into purpose-specific files. It features automated maintenance, including nightly distillation—which condenses daily activity into long-term knowledge—and weekly synthesis that identifies high-level patterns across your interactions.
Central to its functionality is the principle-based enforcement system. P0 to P8 principles are tracked via nightly audits to ensure that tool documentation, decision logic, and failure analysis are consistently captured. This prevents your agent from making recurring mistakes or deferring tasks unnecessarily to you. The skill also offers advanced features such as an AES-256 encrypted vault for secure credentials, opt-in voice profiling for more authentic communication, and automated git backups that automatically scrub sensitive secrets before pushing.
Installation
To install OpenCortex, ensure you are running OpenClaw 2026.2.x+ and have the ClawHub CLI installed. Navigate to your OpenClaw workspace directory (e.g., cd ~/clawd) and run clawhub install opencortex. Once the files are local, execute the installation script using bash skills/opencortex/scripts/install.sh. We highly recommend performing a dry run first (--dry-run) to preview the changes. After installation, you can verify the integrity of the memory architecture by running bash skills/opencortex/scripts/verify.sh. All configurations—including SOUL.md, USER.md, and MEMORY.md—reside locally within your workspace, ensuring total ownership and privacy.
Use Cases
Use OpenCortex when you are starting a fresh project and want your agent to build context from day one, when you notice your agent forgetting key preferences or past decisions, or when you wish to establish a rigorous, principle-based operational framework. It is ideal for developers and power users who require a high degree of memory durability and long-term project continuity. Note: This is not intended for high-frequency runtime memory searches; rely on OpenClaw's built-in memory tools for that.
Example Prompts
- "OpenCortex, please set up memory architecture for my new development project."
- "I've been feeling like you keep forgetting my preferences; can you organize yourself using the OpenCortex protocol?"
- "Run a memory audit and help me refine my core principles to stop you from deferring complex tasks."
Tips & Limitations
OpenCortex prioritizes safety and local-first execution. It makes no network calls during setup and maintains full auditability. Note that features like voice profiling and git auto-pushes are off by default and must be explicitly enabled via environment variables. Treat your .secrets-map file with care to ensure the git scrubbing functionality works as intended. Always customize SOUL.md to reflect the personality and constraints you want your agent to adopt long-term.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-jd2005l-opencortex": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: file-write, file-read, code-execution