Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion .agents/skills/run-eval.md
Original file line number Diff line number Diff line change
Expand Up @@ -32,7 +32,7 @@ curl -X POST \

**Key parameters:**
- `benchmark`: `swebench`, `swebenchmultimodal`, `gaia`, `swtbench`, `commit0`, `multiswebench`
- `eval_limit`: `1`, `50`, `100`, `200`, `500`
- `eval_limit`: Any positive integer (e.g., `1`, `10`, `50`, `200`)
- `model_ids`: See `.github/run-eval/resolve_model_config.py` for available models
- `benchmarks_branch`: Use feature branch from the benchmarks repo to test benchmark changes before merging

Expand Down
18 changes: 10 additions & 8 deletions .github/workflows/run-eval.yml
Original file line number Diff line number Diff line change
Expand Up @@ -32,16 +32,10 @@ on:
default: false
type: boolean
eval_limit:
description: Number of instances to run
description: Number of instances to run (any positive integer)
required: false
default: '1'
type: choice
options:
- '1'
- '100'
- '50'
- '200'
- '500'
type: string
model_ids:
description: Comma-separated model IDs to evaluate. Must be keys of MODELS in resolve_model_config.py. Defaults to first model in that
dict.
Expand Down Expand Up @@ -138,6 +132,14 @@ jobs:
with:
python-version: '3.13'

- name: Validate eval_limit
if: github.event_name == 'workflow_dispatch'
run: |
if ! [[ "${{ github.event.inputs.eval_limit }}" =~ ^[1-9][0-9]*$ ]]; then
echo "Error: eval_limit must be a positive integer, got: ${{ github.event.inputs.eval_limit }}"
exit 1
fi

- name: Validate SDK reference (semantic version check)
if: github.event_name == 'workflow_dispatch'
env:
Expand Down
Loading