bencherdev/bencher

{
  "$schema": "https://agent-plugins.org/schemas/1.0.0/plugin.schema.json",
  "name": "bencher",
  "version": "0.6.13",
  "description": "Catch performance regressions before they merge",
  "author": { "name": "Bencher", "url": "https://bencher.dev" },
  "homepage": "https://bencher.dev",
  "repository": "https://github.com/bencherdev/bencher",
  "keywords": ["benchmarking", "continuous-benchmarking", "performance", "regression-detection", "ci", "bencher"],
  "extensions": {
    "com.openai": {
      "interface": {
        "displayName": "Bencher",
        "shortDescription": "Continuous Benchmarking",
        "longDescription": "Bencher is a continuous benchmarking platform that helps systems software engineers catch performance regressions before they merge.\n\nHow Bencher Works:\n\n1. Run: Run your benchmarks locally or in CI using the exact same bare metal runners and your favorite benchmarking tools. The `bencher` CLI orchestrates running your benchmarks on bare metal and stores the results.\n2. Track: Track the results of your benchmarks over time. Monitor, query, and graph the results using the Bencher web console based on the source branch, testbed, benchmark, and measure.\n3. Catch: Catch performance regressions locally or in CI using the exact same bare metal hardware. Bencher uses state of the art, customizable analytics to detect performance regressions before they merge.\n\nFor the same reasons that unit tests are run to prevent feature regressions, benchmarks should be run with Bencher to prevent performance regressions. Performance bugs are bugs!",
        "developerName": "Bencher",
        "category": "Developer Tools",
        "capabilities": [],
        "websiteURL": "https://bencher.dev",
        "supportURL": "https://bencher.dev/help/",
        "privacyPolicyURL": "https://bencher.dev/legal/privacy/",
        "termsOfServiceURL": "https://bencher.dev/legal/terms-of-use/",
        "defaultPrompt": [
          "Run my benchmarks on Bencher bare metal and show me the results.",
          "Compare my branch to main on Bencher bare metal and flag regressions.",
          "Run my benchmarks on Bencher bare metal in GitHub Actions on every pull request."
        ],
        "brandColor": "#ed6704",
        "brandColorDark": "#ed6704",
        "composerIcon": "./assets/icon.png",
        "logo": "./assets/icon.png"
      },
      "review": {
        "test_cases": {
          "positive": [
            {
              "description": "Show a bare metal run",
              "prompt": "Show the results of the latest bare metal run in the example project on Bencher.",
              "tools_triggered": "list_jobs, view_report",
              "expected_behavior": "Find the newest bare metal job, then show its report with the branch, testbed, and results of each benchmark."
            },
            {
              "description": "View a project",
              "prompt": "Show the details of the example project on Bencher.",
              "tools_triggered": "view_project",
              "expected_behavior": "Return the Example project: name Example, slug example, public visibility, and its repository URL https://github.com/bencherdev/example."
            },
            {
              "description": "List recent reports",
              "prompt": "Show the latest reports for the example project on Bencher.",
              "tools_triggered": "list_reports",
              "expected_behavior": "List the project's most recent reports with the branch, testbed, and time of each one."
            },
            {
              "description": "Query performance history",
              "prompt": "Show the Latency of the bencher::mock_0 benchmark on the main branch and ubuntu-latest testbed of the example project on Bencher over the last 30 days.",
              "tools_triggered": "list_branches, list_testbeds, list_benchmarks, list_measures, query_perf",
              "expected_behavior": "Look up the branch, testbed, benchmark, and measure, then return the Latency value of each run in the last 30 days with its date."
            },
            {
              "description": "Render a performance plot",
              "prompt": "Plot the Latency of the bencher::mock_0 benchmark on the main branch and ubuntu-latest testbed of the example project on Bencher over the last 30 days.",
              "tools_triggered": "list_branches, list_testbeds, list_benchmarks, list_measures, perf_image",
              "expected_behavior": "Look up the branch, testbed, benchmark, and measure, then show a plot image of the Latency values over the last 30 days."
            }
          ],
          "negative": [
            {
              "description": "General coding question unrelated to benchmarking",
              "prompt": "Write a Python function that reverses a string."
            },
            {
              "description": "Unsupported deletion",
              "prompt": "Delete the example project on Bencher."
            },
            {
              "description": "Write to a project the signed-in account is not a member of",
              "prompt": "Create a branch named review-test in the example project on Bencher."
            }
          ]
        }
      },
      "publication": {
        "release_notes": "Initial release with the Bencher MCP tools and the Bencher skill."
      }
    }
  }
}