agent-kernel-evolve
Skill4 starsChanged 3 months agoMITView on GitHub
Proposal-only workflow learning guide for AI agents. Use repeated failures or corrections to draft reviewed, non-persistent learning proposals before any durable write.
1 other project uses a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- imMamdouhaboammar/agent-kerneledited, first seen Jul 26, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 1
- imMamdouhaboammar/agent-kernel★ 547% same
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository15 other files it gives its agents.
- .codex/AGENTS.mdAGENTS.md
- agency-application-security-engineerSkill
- agency-multi-agent-systems-architectSkill
- agency-performance-benchmarkerSkill
- agency-privacy-engineerSkill
- agency-secrets-credential-hygiene-engineerSkill
- agency-test-automation-engineerSkill
- architecture-guardianSkill
- code-reviewSkill
- riqor-agent-orchestratorSkill
- riqor-code-intelligenceSkill
- riqor-session-continuitySkill
- riqor-conventionsSkill
- riqor-verification-gateSkill
- riqor-conventionsSkill
The file
Files
Agent Kernel Self-Evolve Operations Guide
This skill evaluates repeated evidence and drafts a non-persistent proposal by default. It does not authorize automatic learning capture, Playbook creation, Rule creation, hook installation, or durable storage.
Triggers & Trigger Phrases
Activate this skill when:
- The user mentions
self-evolve,/learn,playbook,workflow synthesis, oragent learning. - You resolve a complex multi-step feature or receive a user correction that may justify a future Playbook or Rule. A single event starts proposal evaluation only. Require repeated failures or corrections showing the same pattern before drafting a non-persistent proposal.
1. Playbook Operations (agent-kernel evolve)
Require explicit user approval before any command writes a Playbook, Rule, hook configuration, or durable learning record. Present the repeated evidence, proposed scope, retention, validation cases, rollback, and exact write command before requesting approval. Read-only list and inspect operations may run when already authorized by the task.
- Generate an approved Playbook from a reviewed proposal:
agent-kernel evolve generate --title "Full Next.js Auth Setup" --topic "auth" - List All Generated Playbooks:
agent-kernel evolve list - Inspect a Playbook & Evolution History:
agent-kernel evolve inspect <playbookId> - Evolve/Repair a Playbook after explicit user approval:
agent-kernel evolve repair <playbookId> --reason "Fixed Supabase redirect URL"
2. Universal Self-Evolve Hooks (agent-kernel evolve hooks)
Before either command in this section, obtain explicit opt-in approval for the named agent environments and reviewed paths. Use workflow metadata and synthetic examples only. Reject or redact credentials, personal data, source payloads, command output, and sensitive paths. Define retention, access, and removal controls before enabling durable capture.
- Install Hooks for Antigravity, Claude, Cursor, Codex, Gemini, OpenCode:
agent-kernel evolve hooks - Execute Hook Background Processing after explicit opt-in approval:
agent-kernel hook self-evolve
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

