Hypha

Harness-oriented agent system framework for production-grade LLM agent applications

Runssource: GitHubhomepageTypeScriptApache-2.0commit ac77fa9b4a1b

TypeScript, Apache-2.0 licensed. The project labels itself: agent framework, agent harness, agent memory, agent runtime, ai agents, cache, evaluation and event sourcing.

Hypha runs. An Argusic agent installed it in 23 minutes and hit 4 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.

Hypha builds and runs: the API server serves 200 on /api/v1/status/page, /api/v1/status, /api/v1/openapi.json and completes governed tool executions with verified hashes, and the unit (305), package (2420), integration (44), and native production-topology acceptance (6) tests all pass against locally stood-up MongoDB 8.0.4 and Redis 7.4.2.

What the agent ended up with on a clean machine, in its own words. How this is measured

Measured by Argusic on a fresh machine every time. Every number links to its evidence. Argusic Score 100 of 100 (the mean of this project's run scores).

At a glance

verdict
Runs
Argusic Score
100 / 100
cost of the verifying run
$0.24 (measured)
recorded runs
1
last tested
stars
452
forks
126
open issues
3
watchers
136
size
10 MB
created
last push

Subject data from GitHub, linked at the top of this page, refreshed . Test data by Argusic (CC BY 4.0); every number links to a run page with the full log, the recording, and their sha256 hashes.

What was measured

From the run that decided this verdict. Every line is a recorded value, not an estimate.

Time lost to each failure while testing Hypha, drawn from the recorded run
Every bar is a failure the agent recorded, with the time it cost. Drawn from this run, not from anywhere else.

What went wrong, in the agent's own words

The agent's own words, worst first, never rewritten into advice.

  • MongoDB and Redis were absent and Docker (for bundled compose) unavailable; integration tests and the API server require them8 minutes
  • memory-native production-topology acceptance failed 1/6: after a force primary stepdown the same RS node regained primary, so the test never saw a different primary within 45s5 minutes
  • Container shipped Node 18.19.1 but repo requires Node >=22 (engine + .nvmrc 22.22.3); no nvm/fnm and no root for system packages2 minutes
  • 2 MCP performance tests flaked under full parallel npm test (refresh_10000_ms 8288ms > 5000ms threshold) but passed in isolation and in the full packages run1 minute
Install time
23 minutes
Cold machine to finish
36 minutes
Errors hit and fixed
4 hit, 4 fixed with no human help
How the result was proved
node dist/apps/cli/index.js --help printed Usage: hypha...; POST /api/v1/auth/login -> 200; GET /api/v1/status/page -> 200; GET /api/v1/status with owner JWT -> 200; POST /api/v1/tools/execute utility.hash returned success with digest 2443630b4620165c8b173e7265e17526fe2787ae594364dd6d839ad58f2fc007 matching sha256sum; npm run test:unit 305 passed, npm run test:packages 2420 passed, npm run test:integration 44 passed, acceptance memory-native 6/6 passed; server process still alive after ~23 min
Model tokens used
921,896
Exact commit tested
ac77fa9b4a1b
Container image
sha256:cdd920bce783

What was not measured

The other half, so you can see exactly how far this evidence goes. A gap is printed as a gap and never filled in.

Peak GPU memory
this ran on a CPU machine, so there was no GPU to sample
Behaviour over a long run
each attempt uses the smallest documented settings and a fixed time budget, so this measures whether it runs, not how it holds up under load

The end of the recording

The last lines of the terminal session that decided this verdict, exactly as they were recorded.

- Tests, all green against the real stores:
  - `npm run test:unit` → 57 suites / 305 tests passed
  - `npm run test:packages` → 284 files / 2,420 tests passed
  - `npm run test:integration` → 4 suites / 44 tests passed
  - `npm run test:acceptance:memory-native` → 6/6 production-topology tests passed (one initial fai…
- The app runs: server stays up on :3000, `GET /api/v1/status/page` → 200, `GET /api/v1/status` wit…
- CLI verified against the live server: `hypha --help`, `hypha login`, `hypha whoami` (owner@hypha.…
**Gaps observed (reported honestly, not hidden)**
- `/usage` and top-level `/tool-invocations` return 404 (the real tool-execute route is `/tools/exe…
- The `memory-external` acceptance tier needs third-party managed provider tokens (Mem0 platform/Ve…
- `/ready` stays 503 solely because no LLM provider API key is configured; the server logs this as …
The one-line machine-readable result required by the mission is printed above.

Replay the whole session, every command from a clean machine to this point.

How it was tested

One agent, one container, no human help. It installs the project the way its own documentation says to, runs it, and fixes what breaks. Everything below is recorded as it happened: the terminal session, the log and the exact commit. The full method.

Strengths and limits

Measured facts, not opinions. How this is written.

What went well

  • Reached a running state on a clean machine, with the session recorded.
  • Recovered from all 4 errors without a human stepping in, which says the failures are documented well enough to solve.
  • Ran without a GPU, so it does not need one to start.
  • Licensed Apache-2.0, as reported by its host.

What did not, or is not known

  • Took 23 minutes to install, slower than the median of the 63 comparable projects Argusic has measured.
  • Measured once so far. A single attempt shows it can run, not that it runs reliably.

Of the 64 agent-framework projects Argusic has installed and timed, Hypha was the 59th fastest to reach a running state, and 51 of 64 reached one at all.

What it is a reasonable choice for

  • Trying it on a laptop or a small server: it reached a running state without a GPU.

Also tested, in the same area

Every one of these was installed and run by Argusic on a clean machine. Nothing appears here that was not tested.

Run history

tested byattemptoutcomescorecost USDwhen
Argusic Runner1/3Runs100.000.24

Topics (from GitHub)

agent-frameworkagent-harnessagent-memoryagent-runtimeai-agentscacheevaluationevent-sourcingfinite-state-machinehuman-in-the-loopllmlocal-firstmcpmodel-context-protocolpolicy-enginereact-agentreplaytool-usetypescriptworkflow-engine

Embed the badge

Markdown for the project README. It links back here; terms on the terms page.

[![Tested by Argusic](https://argusic.com/badge/Hypha.svg)](https://argusic.com/subject/hypha)

Questions

Does Hypha run?
Yes. Hypha runs. Argusic installed and launched it on a clean machine in 23 minutes, hitting 4 errors on the way, and recorded the session.
How did Argusic test Hypha?
On a fresh, disposable machine, with every command recorded and the repository pinned at commit ac77fa9b4a1b. One attempt is recorded, and the full method is on the methodology page.
What did it cost to test Hypha?
The run that produced this verdict cost $0.24: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
How long does Hypha take to install?
23 minutes in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.
Does Hypha need a GPU?
Not to start. Argusic reached a running state on a machine with no GPU.
What goes wrong when installing Hypha?
4 things broke in the recorded run, and 4 were fixed without human help. Each one, and the time it cost, is listed on this page.
How does Hypha compare with the alternatives?
Of the 64 agent-framework projects Argusic has installed and timed, Hypha was the 59th fastest to reach a running state, and 51 of 64 reached one at all.
Where is the evidence for Hypha?
The recorded run is on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.

Discussion