From ea0607c48fb92d7d2e9481b4aa69132a97943b7d Mon Sep 17 00:00:00 2001 From: rgarcia <72655+rgarcia@users.noreply.github.com> Date: Tue, 8 Sep 2026 20:17:48 +0000 Subject: [PATCH] Update the pinned ClawBench revision --- .github/workflows/benchmark-clawbench.yml | 4 ++-- benchmarks/harbor/clawbench/run.sh | 2 +- benchmarks/harbor/results.test.ts | 4 ++-- 3 files changed, 5 insertions(+), 5 deletions(-) diff --git a/.github/workflows/benchmark-clawbench.yml b/.github/workflows/benchmark-clawbench.yml index 9f035ab..478340e 100644 --- a/.github/workflows/benchmark-clawbench.yml +++ b/.github/workflows/benchmark-clawbench.yml @@ -284,7 +284,7 @@ jobs: uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4.3.1 with: repository: kernel/ClawBench - ref: c7feaa2435ca8115c0762c44e13885fe5adf3e98 + ref: 9dd9d442c8ec720ba4d6d4c95f2bbc0ff2a2d0df path: clawbench persist-credentials: false @@ -326,7 +326,7 @@ jobs: shell: bash env: CLAWBENCH_REPO: ${{ github.workspace }}/clawbench - CLAWBENCH_REF: c7feaa2435ca8115c0762c44e13885fe5adf3e98 + CLAWBENCH_REF: 9dd9d442c8ec720ba4d6d4c95f2bbc0ff2a2d0df HARBOR_BENCHMARK_TIMEOUT: 4h BENCHMARK_AGENT: ${{ needs.resolve.outputs.agent }} BENCHMARK_TASK: ${{ needs.resolve.outputs.task }} diff --git a/benchmarks/harbor/clawbench/run.sh b/benchmarks/harbor/clawbench/run.sh index 1feda7e..03f9e58 100755 --- a/benchmarks/harbor/clawbench/run.sh +++ b/benchmarks/harbor/clawbench/run.sh @@ -20,7 +20,7 @@ source_root=${KERNEL_MCP_BENCHMARK_SOURCE_ROOT:-$harness_root} benchmark_dir="$harness_root/benchmarks/harbor" image_env="$source_root/benchmarks/harbor/.image.env" clawbench_repo=${CLAWBENCH_REPO:-$harness_root/../ClawBench} -clawbench_ref=${CLAWBENCH_REF:-c7feaa2435ca8115c0762c44e13885fe5adf3e98} +clawbench_ref=${CLAWBENCH_REF:-9dd9d442c8ec720ba4d6d4c95f2bbc0ff2a2d0df} [[ -f "$image_env" ]] || { echo "Missing $image_env; run benchmarks/harbor/build-image.sh first" >&2 diff --git a/benchmarks/harbor/results.test.ts b/benchmarks/harbor/results.test.ts index a077409..3766728 100644 --- a/benchmarks/harbor/results.test.ts +++ b/benchmarks/harbor/results.test.ts @@ -441,7 +441,7 @@ describe("benchmark workflow hardening", () => { expect(workflow).toContain('HARBOR_HYPEMAN_VERSION: "0.1.2"'); expect(workflow).toContain('CODEX_BENCHMARK_VERSION: "0.120.0"'); expect( - workflow.match(/c7feaa2435ca8115c0762c44e13885fe5adf3e98/g), + workflow.match(/9dd9d442c8ec720ba4d6d4c95f2bbc0ff2a2d0df/g), ).toHaveLength(2); expect(workflow).toContain("issues: write\n pull-requests: write"); expect(workflow).not.toContain( @@ -531,7 +531,7 @@ describe("benchmark workflow hardening", () => { ); expect(dockerignore.split("\n")).toContain("*.pem"); expect(runner).not.toContain("KERNEL_PROJECT"); - expect(runner).toContain("c7feaa2435ca8115c0762c44e13885fe5adf3e98"); + expect(runner).toContain("9dd9d442c8ec720ba4d6d4c95f2bbc0ff2a2d0df"); expect(runner).toContain('"${KERNEL_API_BASE_URL%/}/auth/context"'); expect(runner).toContain('bun "$benchmark_dir/verify-project-scope.ts"'); expect(taskPreparer).not.toContain("KERNEL_PROJECT");