gpt-instruct

A Codex jailbreak prompt and test pack for gpt. 针对 gpt 系列的 Codex 破甲提示词与测试包。

Runs with mockssource: GitHubhomepagePythonMITcommit 0ad8ec58e198

Python, MIT licensed.

gpt-instruct runs, with stand-ins for the services it depends on. An Argusic agent installed it in 18 minutes and hit 5 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.

codex-instruct.py deployed gpt-5.6-sol-v45 byte-identically into /home/runner/.codex with model_instructions_file set; 26-unit and 47-scoring tests pass; regression dry-run plans 66 cases; reset removes the managed entry.

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 92 of 100 (the mean of this project's run scores).

At a glance

verdict
Runs with mocks
Argusic Score
92 / 100
cost of the verifying run
$0.04 (measured)
recorded runs
2
last tested
stars
11,501
forks
1,477
open issues
33
watchers
47
size
11 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 gpt-instruct, 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.

  • codex-instruct.py apply wrote to live /home/runner/.codex (CODEX_HOME ignored)6 minutes
  • run_gpt56_sol_issue_regression.py missing default bank and instructions3 minutes
  • scripts/*.zip extraction needed a manual loop; Python extractall filter kwarg unsupported1 minute
  • run_gpt6_astra_project_continuation_probe.py missing /tmp/tests/gpt6_astra_project_continuation_probe.json1 minute
  • python3 -m unittest discover -s unit-tests -q
Install time
18 minutes
Cold machine to finish
3 minutes
Errors hit and fixed
5 hit, 5 fixed with no human help
How the result was proved
python3 -m unittest discover -s unit-tests -q -> OK (26 tests); verify_gpt56_sol_regression_scoring.py -> OK (47 tests); run_gpt56_sol_issue_regression.py --dry-run ... -> 66 cases / 110 turns manifest; zip-entry md sha256 c71c50e2... == deployed /home/runner/.codex/gpt-5.6-sol-v45.md; config.toml gained model_instructions_file line; --reset returned exit 0 and removed the entry/file
Model tokens used
171,245
Exact commit tested
0ad8ec58e198
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
Behaviour against real services
this run was verified against stand-in services only

The end of the recording

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

- Tests: `python3 -m unittest discover -s unit-tests -q` → 26 tests, OK. `verify_gpt56_sol_regressi…
- Toolchain: `run_gpt56_sol_issue_regression.py --dry-run --bank ... --instructions-file ...` print…
- Unfixable gap: the A-probe `run_gpt6_astra_project_continuation_probe.py` needs `tests/gpt6_astra…
- No real credentials needed: the container itself already provides an Argusic-compatible mock at `…
tokens used
171,245
All verification is complete. Summary of what I observed:
- Install: `codex-instruct.py --apply --version gpt-5.6-v45` deployed the prompt into `/home/runner…
- Tests: `python3 -m unittest discover -s unit-tests -q` → 26 tests, OK. `verify_gpt56_sol_regressi…
- Toolchain: `run_gpt56_sol_issue_regression.py --dry-run --bank ... --instructions-file ...` print…
- Unfixable gap: the A-probe `run_gpt6_astra_project_continuation_probe.py` needs `tests/gpt6_astra…
- No real credentials needed: the container itself already provides an Argusic-compatible mock at `…

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

  • Recovered from all 5 errors without a human stepping in, which says the failures are documented well enough to solve.
  • Licensed MIT, as reported by its host.
  • Measured 2 times, so the result is not a one-off.

What did not, or is not known

  • Only reached a running state against stand-in services, so nothing here proves it works against the real ones.
  • Took 18 minutes to install, slower than the median of the 46 comparable projects Argusic has measured.

Of the 47 Python projects Argusic has installed and timed, gpt-instruct was the 47th fastest to reach a running state, and 32 of 47 reached one at all.

Run history

tested byattemptoutcomescorecost USDwhen
Argusic Runner2/3Runs with mocks92.000.04
Argusic Runner1/3Runs with mocks92.000.05

Embed the badge

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

[![Tested by Argusic](https://argusic.com/badge/gpt-instruct.svg)](https://argusic.com/subject/gpt-instruct)

Questions

Does gpt-instruct run?
gpt-instruct runs, with mocks standing in for external services. Argusic installed and launched it on a clean machine in 18 minutes, hitting 5 errors on the way, and recorded the session.
How did Argusic test gpt-instruct?
On a fresh, disposable machine, with every command recorded and the repository pinned at commit 0ad8ec58e198. 2 attempts are recorded, and the full method is on the methodology page.
What did it cost to test gpt-instruct?
The run that produced this verdict cost $0.04: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
How long does gpt-instruct take to install?
18 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 gpt-instruct need a GPU?
Not to start. Argusic reached a running state on a machine with no GPU.
What goes wrong when installing gpt-instruct?
5 things broke in the recorded run, and 5 were fixed without human help. Each one, and the time it cost, is listed on this page.
How does gpt-instruct compare with the alternatives?
Of the 47 Python projects Argusic has installed and timed, gpt-instruct was the 47th fastest to reach a running state, and 32 of 47 reached one at all.
Where is the evidence for gpt-instruct?
All 2 recorded runs are on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.

Discussion