# I don't think you need Claude's harness (slim) > Vlad Cealicu on Anthropic's self-hosted Claude Managed Agents. A company that hosts it already runs the server, the database and the workers, so he thinks it should get Claude to design a harness from its own spec and run its own code, and someone working alone can use Claude Code's cloud sessions. - Full: https://www.anchorterminal.com/blog/another-harness-for-business.md (~2,300 tokens) · this version ~1,230 tokens · JSON https://www.anchorterminal.com/blog/another-harness-for-business.json · canonical https://www.anchorterminal.com/blog/another-harness-for-business - Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-11-13 - Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data) By Vlad Cealicu*, 13 November 2026. 908 words, about 4 min to read. Oh yay, another harness. ## What Anthropic shipped Managed Agents is Anthropic's set of APIs for building and deploying agents. Who runs each part 1. **The loop that calls Claude** (the vendor's code). The harness, which is Anthropic's and which the article calls the brain [[1]](#ref-1) 2. **The server and its database** (you run it). Session history is an event log in your own database [[1]](#ref-1) 3. **The workers and sandboxes** (you run it). Where the agent's shell and file tools run, which the article calls the hands [[1]](#ref-1) 4. **Credentials** (you run it). They can sit in your own secrets manager, kept out of the sandbox [[1]](#ref-1) 5. **Network limits** (you run it). A list of allowed hosts, set in your own firewall [[1]](#ref-1) 6. **The record of every step** (you run it). Exported over OpenTelemetry to the monitoring you already have [[1]](#ref-1) 7. **Requests to Claude** (you pick the route). Anthropic, Amazon Bedrock, Google's Agent Platform, Microsoft Foundry or your own gateway [[1]](#ref-1) So the company already operates all of it, the server, the database and the workers [[1]](#ref-1), and what comes from Anthropic is the code of the harness. ## A harness isn't a secret That list is most of a spec already. None of that is hidden, since we list 56 of them in [agent harnesses](/categories/agent-harnesses) on Anchor Terminal, 25 of them free, and more than half publish their source under MIT or Apache 2.0 [[2]](#ref-2). ## Why your own A harness is where your rules live, which systems an agent may touch, what it has to ask a person before doing, where the log goes and what leaves your machines. I think that matters more at company level, because a company has more rules, and because a harness you own only changes when you change it. ## What you take on You take on the hard part, which is security. You also own the upkeep, so when the model's API changes it's your harness that has to follow. ## If you're solo If you're solo, just use Claude's own harness on the web. ## The other half The other half is the tools the harness calls. If you run a team, write down what an agent may touch, what it has to ask a person and where the log goes, give that to Claude as the spec, and run what comes back on your own machines. --- More in [Opinion](https://www.anchorterminal.com/blog/opinion/index.md) - Older: [What's hyped now, what already peaked, and which company you're in](https://www.anchorterminal.com/blog/what-is-hyped-now.md) - Newer: [A machine that proves Fermat in 29 hours, and the behemoth that's back](https://www.anchorterminal.com/blog/a-machine-that-proves-fermat.md) - All posts: https://www.anchorterminal.com/blog/index.md Anchor Terminal grades the tools AI agents run on, with a source for every score. ## References Read on 13 November 2026. [1] Anthropic, "Operate Claude Managed Agents inside your security perimeter", 12 November 2026. Source for what Managed Agents is, the earlier self-hosted sandbox, the server, database and workers the customer operates, the harness as the brain and the sandbox as the hands, session history in the customer's database, credentials in the customer's secrets manager, the list of allowed hosts set in the customer's firewall, the record of every step exported over OpenTelemetry, where the requests to Claude can go, the private beta and how access is requested. [2] Anchor Terminal, the agent harnesses category, counted on 13 November 2026. Source for the 56 listings, the 25 that are free, the 36 whose licence is MIT or Apache 2.0 in whole or in part, and the four things the category compares harnesses on. [3] Anchor Terminal, the mini-swe-agent listing, checked on 4 November 2026. Source for the agent class of about 190 lines and for CVE-2026-108592 in its experimental bubblewrap sandbox, unfixed at that check. [4] Anchor Terminal, the Claude API listing, which records Anthropic's release note of 7 October 2026. Source for Managed Agents' web fetch tool fetching only URLs already seen in the session from that day, other URLs returning an error, and the beta label. [5] Anchor Terminal, the Claude Code listing, checked on 4 November 2026. Source for the 29 advisories for the Claude Code package in GitHub's reviewed database, all fixed and published. [6] Anthropic, "Use Claude Code in the cloud", Claude Code documentation.