Shareable links

How Cognition compares

One link per question. Short answer up front, honest tables, public sources, and a free trial CTA. Built for posts and DMs, not a sales deck.

Obsidian / LLM wiki

Obsidian and LLM wiki knowledge bases

Obsidian stores notes you curate. LLM wikis retrieve those notes for chat. Cognition turns approved work into skills your coding agents reuse in Cursor, Codex, Claude Code, and Gemini.

Read comparison →

Supermemory

Supermemory

Supermemory is an AI memory API with hybrid search and document ingestion. Cognition is human-approved skill memory for coding agents with triggers, checks, and outcome receipts.

Read comparison →

AGENTS.md / CLAUDE.md

AGENTS.md and CLAUDE.md

AGENTS.md and CLAUDE.md are repo-local instruction files agents read at session start. Cognition adds hosted pedagogy, approved skills, and cross-session memory with outcome receipts.

Read comparison →

Cursor Rules

Cursor Rules

Cursor Rules are project or user instructions inside Cursor. Cognition adds cross-host approved skills, hosted pedagogy, and memory that follows you beyond a single editor config.

Read comparison →

ChatGPT / Claude memory

ChatGPT and Claude built-in memory

ChatGPT Memory and Claude memory keep conversational preferences inside each chat product. Cognition targets coding agents with approved skills, MCP access, and revocable keys.

Read comparison →

Mem0

Mem0

Mem0 is an open source memory layer that extracts and updates facts for personalized agents. Cognition emphasizes human-approved skills, coding-agent installs, and governed reuse receipts.

Read comparison →

Back to Cognition overview