Supermemory remembers for the agents you build. Harbor serves what your team decided to the agents that write your code.
Supermemory is a memory and context engine for the agents and apps you build, with an API, an MCP server and plugins that save and recall coding sessions per repo. Harbor learns from where an engineering team decides, in Slack threads, PR reviews and docs, and serves the rules that apply to each coding session. Supermemory is open source, self-hostable and SOC 2 certified. Harbor has the gate you set, served and cited counts on every rule, and guardrails.
Free for you and one teammate. No card. No end date. Every line about Supermemory is from its own pages, read on 6 November 2026.
Decide what gets in. See what gets used.
Three things Harbor does for the team behind the agents, each beside what Supermemory's own pages say about the same job.
You set the gate
Manual, Hybrid or Auto. Whatever you pick, secrets, install steps and dangerous commands wait for a person.
In SupermemoryCaptures are saved automatically, and the coding plugin documents no review step. Its plugins for Claude Code, Codex, OpenCode and Cursor write to one container per repository.
You see what got used
Every rule shows how often it was served and how often an answer cited it.
In SupermemoryA status command shows the connection and the active project container. The docs don't describe a list of what was recalled.
You block a risky command
Block rules refuse it before it runs, in Claude Code, Cursor, Codex and Copilot.
In SupermemoryNot part of the product.
Harbor and Supermemory, row by row.
| Harbor | Supermemory | |
|---|---|---|
| What it is | Shared memory for a team's coding agents. It learns what your team decides and corrects, and serves the rules that apply to each session. | A memory and context engine for AI agents, available through an API, plugins and an MCP server. Facts live in a graph where a newer fact can update, extend or be derived from older ones. |
| Where the knowledge comes from | Slack, GitHub, GitLab, Notion, Confluence, Jira, Linear, Google Drive and Granola, plus what agent sessions record. | Files, chats, URLs and raw data sent to its API, plus connectors for Google Drive, Gmail, Notion, OneDrive, GitHub, Granola and a web crawler. Its coding plugins save the conversation after each reply. |
| Who decides what gets in | You set the gate. Manual holds everything for a person, Hybrid adds routine rules and holds the important ones, Auto adds everything else. Secrets, install steps and dangerous commands wait in every mode. | Captures are saved automatically, and the coding plugin documents no review step. Its plugins for Claude Code, Codex, OpenCode and Cursor write to one container per repository. |
| How it reaches the agent | Hooks add the rules that apply to the session before the agent writes a line. The same rules are on an MCP server, the CLI and /harbor in Slack. | In Claude Code, a hook loads the project's profile facts when a session starts and searches for matches before each prompt. Agents can also use its MCP server or the API. |
| When a decision changes | When a push deletes the file a rule is about, the rule is retired, or flagged if other repos still use it. A newer rule replaces one it reverses when its author could have retired the old one, with a Restore; otherwise the pair waits for a person. | A new fact replaces the old one for search while the history stays, and temporary facts drop after they expire. |
| Seeing what was used | Every rule shows how often it was served and how often an answer cited it. | A status command shows the connection and the active project container. The docs don't describe a list of what was recalled. |
| Stopping a risky command | Block rules refuse a command before it runs, in Claude Code, Cursor, Codex and GitHub Copilot, and hand the agent something to run instead. | Not part of the product. |
| Agents | Claude Code, Cursor, Codex and GitHub Copilot (the CLI and VS Code). | Plugins for Claude Code, Cursor, Codex, OpenCode, OpenClaw, Hermes and Muse Code, plus ChatGPT, Claude and any MCP client. |
| Open source and hosting | The CLI, harborloop, is MIT licensed on npm. The service is hosted. | Open source under MIT, and runs locally from one binary or Docker. The hosted service self-hosts on Scale, and air-gapped or dedicated on Enterprise. |
| Price | Setup is self-serve and takes two minutes, with no call required. Free costs nothing for up to 2 people on one source, Pro is $24 per person a month billed yearly or $30 monthly, and Premium is $36, or $45 monthly. | Free with $5 of monthly credits, Pro $19 a month, Max $100, Scale $399, and Enterprise is custom. Each plan includes credits, and usage past them is billed. |
| Best for | Engineering teams who want every coding session to start with what the team decided. | Developers building agents and apps that need a memory and context API. |
Supermemory's side comes from its own pages, read on 6 November 2026: supermemory.ai · Supermemory pricing · Supermemory on GitHub · Supermemory in Claude Code · Supermemory connectors · Supermemory graph memory. Out of date? Tell us.
Pick the one built for your job.
Pick Harbor if
- The knowledge is your team's: decisions made in Slack threads and PR reviews, not one user's preferences.
- You want to choose what waits for a person before it reaches every agent.
- You want to see, rule by rule, which ones agents were served and which ones they used.
- You want block rules that stop a destructive command before it runs.
Pick Supermemory if
- You are building an app or agent that needs a memory and search API for its own users.
- You want open source under MIT that you can run on your own machine.
- You need SOC 2, HIPAA or an air-gapped deployment today.
- You use agents beyond Claude Code, Cursor, Codex and Copilot, such as OpenCode, OpenClaw or Hermes.
Before you choose.
Can I run Harbor and Supermemory together?
They can sit side by side. A team can build its product's memory on Supermemory's API and still run Harbor for the agents that write the code. For coding sessions, most teams would pick one.
Can I start with Harbor today?
Yes. Sign up with Google or email, create your workspace and run npx harborloop init, with no card and no call. Supermemory can be signed up for today too.
Is Harbor open source?
The CLI, harborloop, is MIT licensed on npm. The service is hosted. Supermemory: Open source under MIT, and runs locally from one binary or Docker. The hosted service self-hosts on Scale, and air-gapped or dedicated on Enterprise.
Where do the claims about Supermemory come from?
From Supermemory's own public pages, read on 6 November 2026, which are listed under the table. Products move, so if a line is out of date, tell us and we will fix it.
Your agents should know
what your team knows.
Nothing to migrate and nothing to write. Every agent starts with what your team settles, and two weeks in you see which rules they cite.
Free for you and one teammate. No card. No end date.