Cognitive OS v3.0 — Universal AI agent operating system specification. 16-section cognitive architecture for Claude Code, Cursor, Codex, Gemini, and Factory Droid. Multi-harness prompt engineering with verification chain, swarm coordination, and adversarial security.
This repository also configures its own agents. See what cognitive-os tells them →
Skill Claude CodeCodex
Validates that all specification files in the Cognitive OS repo pass markdownlint, are properly formatted, contain required sections, and maintain consistency between master spec and harness adapters.
Skill Claude CodeCodex
Use when a proposed skill, workflow, or decision needs adversarial review before it ships: convenes reviewers drawn from the agent-forge registry as hostile personas, adds a BUYER who states plainly whether they would pay, and forces every verdict to include the cheapest 48-hour test that could kill or validate the…
Skill Claude CodeCodex
Use when it is time for the weekly upstream feed review, or when the user asks what changed in the skill feeds: runs the forge miner diff, clusters new and changed upstream skills by theme, and writes a ranked proposal memo of which patterns are worth authoring into originals.
Skill Claude CodeCodex
Use when installed skills need a lifecycle audit: scans the forge registry and the configured skill directories for staleness, overlap, and conflicting triggers, then proposes a retire, promote, merge, or quarantine action per skill with evidence.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: