Glen records your team's agent sessions. Harbor serves what your team decided.
Glen is shared memory for coding agents. It records your sessions and your team's, connects to apps such as Slack, and adds matching memory to each turn in Claude Code and Codex. 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. Glen adds session takeover, shared transcript search and Pi and Grok Build. 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 Glen is from its own pages, read on 9 November 2026.
Decide what gets in. See what gets used.
Three things Harbor does for the team behind the agents, each beside what Glen'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 GlenEvery finished turn is recorded to one org memory, shared by default, and the docs describe no approval step. Incognito keeps a session out, and access is checked per memory at recall.
You see what got used
Every rule shows how often it was served and how often an answer cited it.
In GlenThe docs don't say how to see which memories were recalled for a turn.
You block a risky command
Block rules refuse it before it runs, in Claude Code, Cursor, Codex and Copilot.
In GlenNot part of the product.
Harbor and Glen, row by row.
| Harbor | Glen | |
|---|---|---|
| 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. | Shared memory for coding agents. It records your sessions and your team's, connects to your apps, and adds context to agents such as Claude Code, Codex and Cursor. |
| Where the knowledge comes from | Slack, GitHub, GitLab, Notion, Confluence, Jira, Linear, Google Drive and Granola, plus what agent sessions record. | Agent sessions from you and your team, with your last 30 days imported on day one, plus connected apps. It reads public material only, such as public Slack channels and shared team transcripts. |
| 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. | Every finished turn is recorded to one org memory, shared by default, and the docs describe no approval step. Incognito keeps a session out, and access is checked per memory at recall. |
| 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 and Codex, hooks add matching team memory on every turn. In Cursor and Grok Build the agent searches with the glen CLI. Slack, the web app and a remote MCP server answer questions too. |
| 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 rule saved by an agent replaces one it reverses when its author could have retired the old one, with a Restore. Otherwise, and for anything read from a connected source, the pair waits for a person. | The docs don't describe a memory being updated or retired on its own. A person can delete one in the dashboard, or ask the agent to use its forget skill. |
| Seeing what was used | Every rule shows how often it was served and how often an answer cited it. | The docs don't say how to see which memories were recalled for a turn. |
| 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). | Claude Code, Codex, Cursor, Pi and Grok Build, plus Claude and ChatGPT through its MCP server. |
| Open source and hosting | The CLI, harborloop, is MIT licensed on npm. The service is hosted. | Its Claude Code plugin is MIT licensed on GitHub. The site does not mention self-hosting. |
| 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 for one person with 15,000 memories, or for a team of up to 3. Personal and Team plans add integrations, and Team has no limits. The site lists no prices. |
| Best for | Engineering teams who want every coding session to start with what the team decided. | Teams who want each agent to recall what teammates' agents already did in past sessions. |
Glen's side comes from its own pages, read on 9 November 2026: tryglen.com · Glen docs · Glen in Claude Code · Glen Claude Code plugin on GitHub. Out of date? Tell us.
Pick the one built for your job.
Pick Harbor if
- You want rules drawn from where the team decides, in Slack threads, PR reviews and docs, not from every agent turn.
- 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 Glen if
- You want every agent turn recorded for the whole team to recall, with nobody writing anything down.
- You want to move a session, yours or a teammate's, to another agent mid-task with its full history.
- You want a shared search over your team's agent transcripts and a library for the reports and specs they produce.
- You use Pi or Grok Build.
Before you choose.
Can I run Harbor and Glen together?
Both give every teammate's agents one shared memory, so most teams would pick one. The split is whether agents recall what past sessions did, or start from the rules the team decided, with a gate on what gets in.
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. Glen can be signed up for today too.
Is Harbor open source?
The CLI, harborloop, is MIT licensed on npm. The service is hosted. Glen: Its Claude Code plugin is MIT licensed on GitHub. The site does not mention self-hosting.
Where do the claims about Glen come from?
From Glen's own public pages, read on 9 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 team’s rules,
in every agent, checked on every PR.
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.