This PR adds support for the eval to read environment variables from a
`.env` file located in the `crates/eval` directory.
For instance, you can use it to set your Anthropic API key:
```
ANTHROPIC_API_KEY=<secret>
```
Release Notes:
- N/A
This update generates a single self-contained .html file that shows an
overview of evaluation threads in the browser. It's useful for:
- Quickly reviewing results
- Sharing evaluation runs
- Debugging
- Comparing models (TBD)
Features:
- Export thread JSON from the UI
- Keyboard navigation (j/k or Ctrl + ←/→)
- Toggle between compact and full views
Generating the overview:
- `cargo run -p eval` will write this file in the run dir's root.
- Or you can call `cargo run -p eval --bin explorer` to generate it
without running evals.
Screenshot:
Release Notes:
- N/A
### Problem
We want to start continuously tracking our progress on agent evals over
time. As part of this, we'd like the *score* to have a clear,
interpretable meaning. Right now, it's a number from 0 to 5, but it's
not clear what any particular number works. In addition, scores vary
widely from run to run, because the agent's output is deterministic. We
try to stabilize the score using a panel of judges, but the behavior of
the agent itself varies much more widely than the judges' scores for a
given run.
### Solution
* **explicit meanings of scores** - In this PR, we're prescribing the
diff and thread criteria files so that they *must* be unordered lists of
assertions. For both the thread and the diff, rather than providing an
abstract score, the judge's task is simply to count how many of these
assertions are satisfied. A percentage score can be derived from this
number, divided by the total number of assertions.
* **repetitions** - Rather than running each example once, and judging
it N times, we'll **run** the example N times. Right now, I'm just
judging the output once per run, because I believe that with these more
clear scoring criteria, the main source of non-determinism will be the
*agent's* behavior, not the judge's
### Questions
* **accounting for diagnostic errors** - Previously, the judge was asked
to incorporate diagnostics into their abstract scores. Now that the
"score" is determined directly from the criteria, the diagnostic will
not be captured in the score. How should the diagnostics be accounted
for in the eval? One thought is - let's simply count and report the
number of errors remaining after the agent finishes, as a separate field
of the run (along with diff score and thread score). We could consider
normalizing it using the total lines of added code (like errors per 100
lines of code added) in order to give it some semblance of stability
between examples.
* **repetitions** - How many repetitions should we run on CI? Each
repetition takes significant time, but I think running more than one
repetition will make the scores significantly less volatile.
### Todo
* [x] Fix `--concurrency` implementation so that only N tasks are
spawned
* [x] Support `--repetitions` efficiently (re-using the same worktree)
* [x] Restructure judge prompts to count passing criteria, not compute
abstract score
* [x] Report total number of diagnostics in some way
* [x] Format output nicely
Release Notes:
- N/A *or* Added/Fixed/Improved ...
---------
Co-authored-by: Antonio Scandurra <me@as-cii.com>
These are only written when the diff has changed. `patch.diff` has been
removed as its redundant with `last.diff`.
It can be convenient to open `last.diff` and use undo/redo to navigate
its history.
Release Notes:
- N/A
Split `locator` out of DebugTaskDefinition to make it clearer when
location needs to happen.
Release Notes:
- N/A
---------
Co-authored-by: Anthony Eid <hello@anthonyeid.me>
Co-authored-by: Anthony <anthony@zed.dev>
Co-authored-by: Cole Miller <m@cole-miller.net>
Now that we've established a proper eval in tree, this PR is reboots of
our agent loop back to a set of minimal tools and simpler prompts. We
should aim to get this branch feeling subjectively competitive with
what's on main and then merge it, and build from there.
Let's invest in our eval and use it to drive better performance of the
agent loop. How you can help: Pick an example, and then make the outcome
faster or better. It's fine to even use your own subjective judgment, as
our evaluation criteria likely need tuning as well at this point. Focus
on making the agent work better in your own subjective experience first.
Let's focus on simple/practical improvements to make this thing work
better, then determine how we can craft our judgment criteria to lock
those improvements in.
Release Notes:
- N/A
---------
Co-authored-by: Max <max@zed.dev>
Co-authored-by: Antonio <antonio@zed.dev>
Co-authored-by: Agus <agus@zed.dev>
Co-authored-by: Richard <richard@zed.dev>
Co-authored-by: Max Brunsfeld <maxbrunsfeld@gmail.com>
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Co-authored-by: Michael Sloan <mgsloan@gmail.com>
The `always_allow_tool_actions` setting would get overridden with the
default when we loaded each example project, leading to examples
stalling when they run a tool that needed confirmation. There's now a
separate `runner_settings.json` file where we can configure the
environment for the eval.
Release Notes:
- N/A
---------
Co-authored-by: Oleksiy <oleksiy@zed.dev>
Release Notes:
- Fixed a regression that caused the agent to hang sometimes.
---------
Co-authored-by: Thomas Mickley-Doyle <tmickleydoyle@gmail.com>
Co-authored-by: Nathan Sobo <nathan@zed.dev>
Co-authored-by: Michael Sloan <mgsloan@gmail.com>
Also, we moved the logic for driving the agentic loop into `Thread` so
that we don't have to re-implement it.
Release Notes:
- N/A
---------
Co-authored-by: Nathan Sobo <nathan@zed.dev>
Richard Feldman
Marshall Bowers
Oleksiy Syvokon
Agus ZubiagaandRichard Feldman
Michael Sloan
Thomas Mickley-DoyleandJoseph