Add programbench to run-eval benchmark choices

.agents/skills/manage-evals/SKILL.md

@@ -69,6 +69,7 @@ done
 | `commit0` | Commit0 — commit generation tasks |
 | `swebenchmultimodal` | SWE-bench Multimodal — tasks with images |
 | `terminalbench` | TerminalBench — terminal interaction tasks |
+| `programbench` | ProgramBench — program-repair tasks against gold-standard test binaries |
 
 ### Trigger Options
 
@@ -129,7 +130,7 @@ Each line is a run path. Match by benchmark and model to find the run.
 ### Step 2: Identify the Run Path Components
 
 A run path has three components:
-- **benchmark**: `swebench`, `gaia`, `swtbench`, `commit0`, `swebenchmultimodal`, `terminalbench`
+- **benchmark**: `swebench`, `gaia`, `swtbench`, `commit0`, `swebenchmultimodal`, `terminalbench`, `programbench`
 - **model_slug**: Derived from model name with `/:@.` replaced by `-` (e.g., `litellm_proxy-claude-sonnet-4-5-20250929`)
 - **run_id**: The GitHub Actions workflow run ID from the `OpenHands/evaluation` repo
 

.agents/skills/manage-evals/references/eval-infrastructure.md

@@ -42,7 +42,7 @@ and served via CDN at `https://results.eval.all-hands.dev/`.
 {benchmark}/{model_slug}/{github_run_id}/
 ```
 
-- **benchmark**: `swebench`, `gaia`, `swtbench`, `commit0`, `swebenchmultimodal`, `terminalbench`
+- **benchmark**: `swebench`, `gaia`, `swtbench`, `commit0`, `swebenchmultimodal`, `terminalbench`, `programbench`
 - **model_slug**: Model name with `/:@.` replaced by `-`
   - Example: `litellm_proxy/claude-sonnet-4-5-20250929` → `litellm_proxy-claude-sonnet-4-5-20250929`
 - **github_run_id**: The GitHub Actions run ID from the `OpenHands/evaluation` repo

.agents/skills/manage-evals/scripts/manage_evals.py

@@ -42,6 +42,7 @@
     "commit0",
     "swebenchmultimodal",
     "terminalbench",
+    "programbench",
 ]
 TOOL_PRESETS = ["default", "gemini", "gpt5", "planning"]
 AGENT_TYPES = ["default", "acp-claude", "acp-codex"]

.agents/skills/run-eval.md

@@ -31,7 +31,7 @@ curl -X POST \
 ```
 
 **Key parameters:**
-- `benchmark`: `swebench`, `swebenchmultimodal`, `gaia`, `swtbench`, `commit0`, `multiswebench`, `terminalbench`
+- `benchmark`: `swebench`, `swebenchmultimodal`, `gaia`, `swtbench`, `commit0`, `multiswebench`, `terminalbench`, `programbench`
 - `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

.github/workflows/run-eval.yml

@@ -21,6 +21,7 @@ on:
                     - commit0
                     - swebenchmultimodal
                     - terminalbench
+                    - programbench
             sdk_ref:
                 description: SDK commit/ref to evaluate (must be a semantic version like v1.0.0 unless 'Allow unreleased branches' is checked)
                 required: true