657 projects tested, and "Rust toolchain not installed" stopped 8: week of September 28, 2026
657 projects installed and run on clean machines in the week of September 28, 2026, 441 verified against the real thing. Every number links to a recorded run.
Covering 2026-09-28 to 2026-10-05. Published .
What was measured
- Projects run
- 657 over 787 attempts (468 the week before)
- Verified against the real thing
- 441
- Verified with stand-ins only
- 159
- Could not be verified
- 187
- Tested here for the first time
- 656
- What it cost us
- 197.17 USD, 34% of it model tokens
What it ran on
Argusic publishes other people's server prices every day. These are its own: the machines this week's work was rented on, at the rates actually paid, each one destroyed when its job finished.
- Attempts on CPU machines
- 805 at 0.0997 USD an hour
- Data centers used
- EU-CZ-1, EU-NL-1, EU-RO-1, EU-SE-1, EUR-IS-1, US-CA-2, US-GA-2, US-KS-2, US-MD-1, US-MO-1, US-NC-1, US-NC-2, US-NE-1
What the week cost, broken down
A single total answers nothing for somebody working out whether they could do this themselves. The difference between the dearest and the cheapest project is how much fixing each needed, not a price anybody chose.
- Model tokens
- 66.06 USD
- Machines
- 131.11 USD
- Per project measured
- 0.30 USD
- Dearest project
- LumixEngine at 1.35 USD
- Cheapest project
- alive-progress at 0.00 USD
What the pool looks like now
The state of the whole body of evidence at the end of this week, not only what the week itself added.
- Projects measured in total
- 1409
- Of those, reached a working state
- 1031
- Most measured languages
- Python (371), TypeScript (314), Go (202), Rust (153), JavaScript (135)
What kept breaking
Counted across projects, not across attempts, so one repository retried three times is one problem. Each label is the wording an agent actually used.

- 8 projects
- Rust toolchain not installed
- 6 projects
- Go compiler not found in container
- 6 projects
- pnpm not found in PATH
- 4 projects
- No Go compiler in container
- 3 projects
- 1 test failure
The result most worth reading
nanoc scored 100 of 100, after hitting 9 errors that the agent fixed with no human help.
Nanoc 4.14.8 loads and its core test suite (59 tests across base/ and orig_cli/) passes with 0 failures and 0 errors after installing Ruby 3.2 via extracted Ubuntu debs, setting up dev headers and linker symlinks, providing a libyaml shared library, and creating an assert shim for the rbtree native extension.
The agent's own words, from its recorded report.
What this week does not tell you
- Behavior over time
- each attempt uses the smallest documented settings and a fixed time budget, so this measures whether a thing runs, not how it holds up under load
- Anything not tested
- only the 657 projects above were run this week; a project absent from this report was not measured, not judged
- Real-service behavior for 159 of them
- those ran against stand-in services, never the real ones
The report as written
Argusic installed and ran 657 projects on clean machines between 2026-09-28 and 2026-10-05, over 787 attempts, 189 more subjects than the week before.
441 were verified against the real thing, 159 only against stand-in services, and 187 could not be verified at all.
656 projects were tested here for the first time.
The same problems keep appearing across unrelated projects. The most common this week was "Rust toolchain not installed", which stopped 8 projects. Also recurring: "Go compiler not found in container" (6), "pnpm not found in PATH" (6), "No Go compiler in container" (4), "1 test failure" (3).
The result most worth reading was nanoc, which scored 100 of 100 after hitting 9 errors that the agent fixed with no human help. In the agent's own words: Nanoc 4.14.8 loads and its core test suite (59 tests across base/ and orig_cli/) passes with 0 failures and 0 errors after installing Ruby 3.2 via extracted Ubuntu debs, setting up dev headers and linker symlinks, providing a libyaml shared library, and creating an assert shim for the rbtree native extension.
Running all of this cost 197.17 USD, of which 34% was model tokens rather than machines.
66.06 USD of that was model tokens and 131.11 USD was machines, which is 0.30 USD for each project measured.
The dearest project of the week was LumixEngine at 1.35 USD, the cheapest alive-progress at 0.00 USD. The difference is how much fixing each one needed, not a price we chose.
The work ran on machines rented by the hour and destroyed afterwards: 805 attempts on CPU machines at 0.0997 USD an hour. Those are the rates Argusic actually paid, not list prices. The pods came up in 13 data centers: EU-CZ-1, EU-NL-1, EU-RO-1, EU-SE-1, EUR-IS-1, US-CA-2, US-GA-2, US-KS-2, US-MD-1, US-MO-1, US-NC-1, US-NC-2, US-NE-1.
Across everything measured so far, 1409 projects have been installed and run here, and 1031 of them reached a working state. Most of them are written in Python (371), TypeScript (314), Go (202), Rust (153), JavaScript (135).
Every number above comes from a recorded run, and each one has its log and its terminal recording on this site.