diff --git a/package-lock.json b/package-lock.json
index 8d62a9c75..10f4b3772 100644
--- a/package-lock.json
+++ b/package-lock.json
@@ -794,6 +794,10 @@
 			"resolved": "packages/coding-agent",
 			"link": true
 		},
+		"node_modules/@earendil-works/pi-evals": {
+			"resolved": "packages/evals",
+			"link": true
+		},
 		"node_modules/@earendil-works/pi-server": {
 			"resolved": "packages/server",
 			"link": true
@@ -1479,6 +1483,52 @@
 				"node": ">= 10"
 			}
 		},
+		"node_modules/@mariozechner/pi-agent-core": {
+			"name": "@earendil-works/pi-agent-core",
+			"version": "0.81.1",
+			"resolved": "https://registry.npmjs.org/@earendil-works/pi-agent-core/-/pi-agent-core-0.81.1.tgz",
+			"integrity": "sha512-yqbh68CyhqxMov/jUogFJfMqlu2Gd37GAki+tr59YCmAPHfomiCA5ESzusXtpGzABeiZFC/OrRdQ4GwCCOMIHA==",
+			"dev": true,
+			"license": "MIT",
+			"peer": true,
+			"dependencies": {
+				"@earendil-works/pi-ai": "^0.81.1",
+				"ignore": "7.0.5",
+				"typebox": "1.1.38",
+				"yaml": "2.9.0"
+			},
+			"engines": {
+				"node": ">=22.19.0"
+			}
+		},
+		"node_modules/@mariozechner/pi-ai": {
+			"name": "@earendil-works/pi-ai",
+			"version": "0.81.1",
+			"resolved": "https://registry.npmjs.org/@earendil-works/pi-ai/-/pi-ai-0.81.1.tgz",
+			"integrity": "sha512-hzHE7Z8l5mgJk+ke67Lge0rwS2+wbKJrFKl9o5M1R1rh33+cCT7D1AHz1OAtX5wFs90E1/BTGhyJRTUHaMxGvQ==",
+			"dev": true,
+			"license": "MIT",
+			"peer": true,
+			"dependencies": {
+				"@anthropic-ai/sdk": "0.91.1",
+				"@aws-sdk/client-bedrock-runtime": "3.1048.0",
+				"@google/genai": "1.52.0",
+				"@mistralai/mistralai": "2.2.6",
+				"@opentelemetry/api": "1.9.0",
+				"@smithy/node-http-handler": "4.7.3",
+				"http-proxy-agent": "7.0.2",
+				"https-proxy-agent": "7.0.6",
+				"openai": "6.26.0",
+				"partial-json": "0.1.7",
+				"typebox": "1.1.38"
+			},
+			"bin": {
+				"pi-ai": "dist/cli.js"
+			},
+			"engines": {
+				"node": ">=22.19.0"
+			}
+		},
 		"node_modules/@mistralai/mistralai": {
 			"version": "2.2.6",
 			"resolved": "https://registry.npmjs.org/@mistralai/mistralai/-/mistralai-2.2.6.tgz",
@@ -2288,6 +2338,151 @@
 				"win32"
 			]
 		},
+		"node_modules/@vitest-evals/core": {
+			"version": "0.15.0",
+			"resolved": "https://registry.npmjs.org/@vitest-evals/core/-/core-0.15.0.tgz",
+			"integrity": "sha512-kqyg3ocVtZxr3rP9oh01oAycRgXNRasvTgLZ8MbFGBJ4BRd4tk7Bgww253HY7fw7ZFUp4UPYNmDp5AcHajm++w==",
+			"dev": true,
+			"license": "Apache-2.0",
+			"dependencies": {
+				"zod": ">=3 <5"
+			}
+		},
+		"node_modules/@vitest-evals/harness-pi-ai": {
+			"version": "0.15.0",
+			"resolved": "https://registry.npmjs.org/@vitest-evals/harness-pi-ai/-/harness-pi-ai-0.15.0.tgz",
+			"integrity": "sha512-iigLAFAZdJ4wefndL2pkX69lJ682QisN8dNhXx/rsJQGTsjXA0dQekw5uww8CCwOIJVagyII8y9XOmdJutz73Q==",
+			"dev": true,
+			"license": "Apache-2.0",
+			"peerDependencies": {
+				"@mariozechner/pi-agent-core": ">=0.67 <1",
+				"@mariozechner/pi-ai": ">=0.67 <1",
+				"vitest-evals": "*"
+			}
+		},
+		"node_modules/@vitest-evals/report-ui": {
+			"version": "0.15.0",
+			"resolved": "https://registry.npmjs.org/@vitest-evals/report-ui/-/report-ui-0.15.0.tgz",
+			"integrity": "sha512-icrutiRhZ2T1MrHo2RW09hcEwfi/tV/FgsGAE76Wd0C02tZyCleQRjt11RWhBJiYDPpsjjjTHGCHIO6tmtX5zQ==",
+			"dev": true,
+			"license": "Apache-2.0",
+			"dependencies": {
+				"@vitest-evals/core": "0.15.0"
+			}
+		},
+		"node_modules/@vitest/expect": {
+			"version": "4.1.9",
+			"resolved": "https://registry.npmjs.org/@vitest/expect/-/expect-4.1.9.tgz",
+			"integrity": "sha512-vl/rYsUKcBr3SnQn166+XR5ZQcgMx3DQhFWdfli/cWpLnLUmbxZvyrJZotLFUryib+LtArYMSTJ5RbQ57ZqrlA==",
+			"dev": true,
+			"license": "MIT",
+			"dependencies": {
+				"@standard-schema/spec": "^1.1.0",
+				"@types/chai": "^5.2.2",
+				"@vitest/spy": "4.1.9",
+				"@vitest/utils": "4.1.9",
+				"chai": "^6.2.2",
+				"tinyrainbow": "^3.1.0"
+			},
+			"funding": {
+				"url": "https://opencollective.com/vitest"
+			}
+		},
+		"node_modules/@vitest/mocker": {
+			"version": "4.1.9",
+			"resolved": "https://registry.npmjs.org/@vitest/mocker/-/mocker-4.1.9.tgz",
+			"integrity": "sha512-EVkXzBjrPGM+cK8/ANWgBrkUCfJfb38/EfTSO8h7pWvKkyPkpWxvR7BkD2MyItMF62C97zAEoqdpUixwR/e+Rw==",
+			"dev": true,
+			"license": "MIT",
+			"dependencies": {
+				"@vitest/spy": "4.1.9",
+				"estree-walker": "^3.0.3",
+				"magic-string": "^0.30.21"
+			},
+			"funding": {
+				"url": "https://opencollective.com/vitest"
+			},
+			"peerDependencies": {
+				"msw": "^2.4.9",
+				"vite": "^6.0.0 || ^7.0.0 || ^8.0.0"
+			},
+			"peerDependenciesMeta": {
+				"msw": {
+					"optional": true
+				},
+				"vite": {
+					"optional": true
+				}
+			}
+		},
+		"node_modules/@vitest/pretty-format": {
+			"version": "4.1.9",
+			"resolved": "https://registry.npmjs.org/@vitest/pretty-format/-/pretty-format-4.1.9.tgz",
+			"integrity": "sha512-s0iufns3iIFitdgm+YR7g1whCAaGtXz459VS9/PqyKDEEFgYIhsHOQmXgIgDuYCt7DeQmiZT0Qe2OA2p4ZPu5A==",
+			"dev": true,
+			"license": "MIT",
+			"dependencies": {
+				"tinyrainbow": "^3.1.0"
+			},
+			"funding": {
+				"url": "https://opencollective.com/vitest"
+			}
+		},
+		"node_modules/@vitest/runner": {
+			"version": "4.1.9",
+			"resolved": "https://registry.npmjs.org/@vitest/runner/-/runner-4.1.9.tgz",
+			"integrity": "sha512-KXLMDtc7oe70+3mJfGrPUWPesswH+3sTxAMAMl8DG7I8IUQT4XW718dY5ID3vPUcmlu27CcKfY4P3h3I29SLJg==",
+			"dev": true,
+			"license": "MIT",
+			"dependencies": {
+				"@vitest/utils": "4.1.9",
+				"pathe": "^2.0.3"
+			},
+			"funding": {
+				"url": "https://opencollective.com/vitest"
+			}
+		},
+		"node_modules/@vitest/snapshot": {
+			"version": "4.1.9",
+			"resolved": "https://registry.npmjs.org/@vitest/snapshot/-/snapshot-4.1.9.tgz",
+			"integrity": "sha512-Jc7RKGNBo8Z28WYIm0Niej4xdSPByRf6mU58VpHQkd6Zh05rlnA+twjbK5HyeIGHxrzsc3mJgS43uM0CZKzaIA==",
+			"dev": true,
+			"license": "MIT",
+			"dependencies": {
+				"@vitest/pretty-format": "4.1.9",
+				"@vitest/utils": "4.1.9",
+				"magic-string": "^0.30.21",
+				"pathe": "^2.0.3"
+			},
+			"funding": {
+				"url": "https://opencollective.com/vitest"
+			}
+		},
+		"node_modules/@vitest/spy": {
+			"version": "4.1.9",
+			"resolved": "https://registry.npmjs.org/@vitest/spy/-/spy-4.1.9.tgz",
+			"integrity": "sha512-fHpsS6mIi+PiEW+vcRVOMkX1oSaPKne3VOclSFICPcGOmfKgXPU5iAah+wcNcj2xPrCCmfq99IDGf+EojhhvhA==",
+			"dev": true,
+			"license": "MIT",
+			"funding": {
+				"url": "https://opencollective.com/vitest"
+			}
+		},
+		"node_modules/@vitest/utils": {
+			"version": "4.1.9",
+			"resolved": "https://registry.npmjs.org/@vitest/utils/-/utils-4.1.9.tgz",
+			"integrity": "sha512-A51o8ymO5PpqlWNnBP9ZHPXDIpuMtTLlGSjN7la4US+LJzoUMyhwjA5QXlm39JexgwHKW4Xjs8Z2d3dLCXOeuA==",
+			"dev": true,
+			"license": "MIT",
+			"dependencies": {
+				"@vitest/pretty-format": "4.1.9",
+				"convert-source-map": "^2.0.0",
+				"tinyrainbow": "^3.1.0"
+			},
+			"funding": {
+				"url": "https://opencollective.com/vitest"
+			}
+		},
 		"node_modules/@xterm/headless": {
 			"version": "5.5.0",
 			"resolved": "https://registry.npmjs.org/@xterm/headless/-/headless-5.5.0.tgz",
@@ -2480,6 +2675,16 @@
 				"node": ">=20"
 			}
 		},
+		"node_modules/chai": {
+			"version": "6.2.2",
+			"resolved": "https://registry.npmjs.org/chai/-/chai-6.2.2.tgz",
+			"integrity": "sha512-NUPRluOfOiTKBKvWPtSD4PhFvWCqOi0BGStNWs57X9js7XGTprSmFoz5F0tWhR4WPjNeR9jXqdC7/UpSJTnlRg==",
+			"dev": true,
+			"license": "MIT",
+			"engines": {
+				"node": ">=18"
+			}
+		},
 		"node_modules/chalk": {
 			"version": "5.6.2",
 			"resolved": "https://registry.npmjs.org/chalk/-/chalk-5.6.2.tgz",
@@ -2643,6 +2848,13 @@
 				"node": ">= 0.4"
 			}
 		},
+		"node_modules/es-module-lexer": {
+			"version": "2.3.1",
+			"resolved": "https://registry.npmjs.org/es-module-lexer/-/es-module-lexer-2.3.1.tgz",
+			"integrity": "sha512-shc1dbU90Yl/xq1QrC7QRtfcwURZuVRfPhZbDoldJ1cn1gzDvBaBWlv0eFolj5+0znnPJz5TXLxsN77X/12KTA==",
+			"dev": true,
+			"license": "MIT"
+		},
 		"node_modules/esbuild": {
 			"version": "0.28.1",
 			"resolved": "https://registry.npmjs.org/esbuild/-/esbuild-0.28.1.tgz",
@@ -4586,6 +4798,13 @@
 			"dev": true,
 			"license": "MIT"
 		},
+		"node_modules/std-env": {
+			"version": "4.2.0",
+			"resolved": "https://registry.npmjs.org/std-env/-/std-env-4.2.0.tgz",
+			"integrity": "sha512-oCUKSupKTHX53EyjDtuZQ64pjLJ6yYCtpmEw0goYxtjG9KpbRe8KAsl2tBUGU9DyMcJ0RwJ8GqJAFzMXcXW1Rw==",
+			"dev": true,
+			"license": "MIT"
+		},
 		"node_modules/string_decoder": {
 			"version": "1.3.0",
 			"resolved": "https://registry.npmjs.org/string_decoder/-/string_decoder-1.3.0.tgz",
@@ -4678,6 +4897,16 @@
 			"dev": true,
 			"license": "MIT"
 		},
+		"node_modules/tinyexec": {
+			"version": "1.2.4",
+			"resolved": "https://registry.npmjs.org/tinyexec/-/tinyexec-1.2.4.tgz",
+			"integrity": "sha512-SHf/r48b7vOrjve9PxJo3MN5v5yuyjHvdUcrQffT3WXMUfnGmHDVbC4k3sHJaJTgZCwpUplIaAo5ANtMyp3YHg==",
+			"dev": true,
+			"license": "MIT",
+			"engines": {
+				"node": ">=18"
+			}
+		},
 		"node_modules/tinyglobby": {
 			"version": "0.2.17",
 			"resolved": "https://registry.npmjs.org/tinyglobby/-/tinyglobby-0.2.17.tgz",
@@ -4726,6 +4955,16 @@
 				"url": "https://github.com/sponsors/jonschlinkert"
 			}
 		},
+		"node_modules/tinyrainbow": {
+			"version": "3.1.0",
+			"resolved": "https://registry.npmjs.org/tinyrainbow/-/tinyrainbow-3.1.0.tgz",
+			"integrity": "sha512-Bf+ILmBgretUrdJxzXM0SgXLZ3XfiaUuOj/IKQHuTXip+05Xn+uyEYdVg0kYDipTBcLrCVyUzAPz7QmArb0mmw==",
+			"dev": true,
+			"license": "MIT",
+			"engines": {
+				"node": ">=14.0.0"
+			}
+		},
 		"node_modules/to-regex-range": {
 			"version": "5.0.1",
 			"resolved": "https://registry.npmjs.org/to-regex-range/-/to-regex-range-5.0.1.tgz",
@@ -4922,6 +5161,137 @@
 				"url": "https://github.com/sponsors/jonschlinkert"
 			}
 		},
+		"node_modules/vitest": {
+			"version": "4.1.9",
+			"resolved": "https://registry.npmjs.org/vitest/-/vitest-4.1.9.tgz",
+			"integrity": "sha512-nE3/LEyc0z87uHYLZebqCUOaJr2hdtuPp7BQ4BosVFnfltxgAvMG08NyrSGlPpOUWvR27c5flSmYFTNr78L9GQ==",
+			"dev": true,
+			"license": "MIT",
+			"dependencies": {
+				"@vitest/expect": "4.1.9",
+				"@vitest/mocker": "4.1.9",
+				"@vitest/pretty-format": "4.1.9",
+				"@vitest/runner": "4.1.9",
+				"@vitest/snapshot": "4.1.9",
+				"@vitest/spy": "4.1.9",
+				"@vitest/utils": "4.1.9",
+				"es-module-lexer": "^2.0.0",
+				"expect-type": "^1.3.0",
+				"magic-string": "^0.30.21",
+				"obug": "^2.1.1",
+				"pathe": "^2.0.3",
+				"picomatch": "^4.0.3",
+				"std-env": "^4.0.0-rc.1",
+				"tinybench": "^2.9.0",
+				"tinyexec": "^1.0.2",
+				"tinyglobby": "^0.2.15",
+				"tinyrainbow": "^3.1.0",
+				"vite": "^6.0.0 || ^7.0.0 || ^8.0.0",
+				"why-is-node-running": "^2.3.0"
+			},
+			"bin": {
+				"vitest": "vitest.mjs"
+			},
+			"engines": {
+				"node": "^20.0.0 || ^22.0.0 || >=24.0.0"
+			},
+			"funding": {
+				"url": "https://opencollective.com/vitest"
+			},
+			"peerDependencies": {
+				"@edge-runtime/vm": "*",
+				"@opentelemetry/api": "^1.9.0",
+				"@types/node": "^20.0.0 || ^22.0.0 || >=24.0.0",
+				"@vitest/browser-playwright": "4.1.9",
+				"@vitest/browser-preview": "4.1.9",
+				"@vitest/browser-webdriverio": "4.1.9",
+				"@vitest/coverage-istanbul": "4.1.9",
+				"@vitest/coverage-v8": "4.1.9",
+				"@vitest/ui": "4.1.9",
+				"happy-dom": "*",
+				"jsdom": "*",
+				"vite": "^6.0.0 || ^7.0.0 || ^8.0.0"
+			},
+			"peerDependenciesMeta": {
+				"@edge-runtime/vm": {
+					"optional": true
+				},
+				"@opentelemetry/api": {
+					"optional": true
+				},
+				"@types/node": {
+					"optional": true
+				},
+				"@vitest/browser-playwright": {
+					"optional": true
+				},
+				"@vitest/browser-preview": {
+					"optional": true
+				},
+				"@vitest/browser-webdriverio": {
+					"optional": true
+				},
+				"@vitest/coverage-istanbul": {
+					"optional": true
+				},
+				"@vitest/coverage-v8": {
+					"optional": true
+				},
+				"@vitest/ui": {
+					"optional": true
+				},
+				"happy-dom": {
+					"optional": true
+				},
+				"jsdom": {
+					"optional": true
+				},
+				"vite": {
+					"optional": false
+				}
+			}
+		},
+		"node_modules/vitest-evals": {
+			"version": "0.15.0",
+			"resolved": "https://registry.npmjs.org/vitest-evals/-/vitest-evals-0.15.0.tgz",
+			"integrity": "sha512-bOYALQdLXKYFRCV8aQ+yLeH7XkmYsTxUWQxLHzJdIEcIBx2v/kPgG3Zg6m231jMCphuoM6xH4YQB3+0pDP7LLw==",
+			"dev": true,
+			"license": "Apache-2.0",
+			"dependencies": {
+				"@vitest-evals/core": "0.15.0",
+				"@vitest-evals/report-ui": "0.15.0"
+			},
+			"bin": {
+				"vitest-evals": "bin/vitest-evals.js"
+			},
+			"peerDependencies": {
+				"ai": ">=4 <7",
+				"tinyrainbow": ">=2 <4",
+				"vitest": ">=4 <5",
+				"zod": ">=3 <5"
+			},
+			"peerDependenciesMeta": {
+				"ai": {
+					"optional": true
+				},
+				"zod": {
+					"optional": true
+				}
+			}
+		},
+		"node_modules/vitest/node_modules/picomatch": {
+			"version": "4.0.5",
+			"resolved": "https://registry.npmjs.org/picomatch/-/picomatch-4.0.5.tgz",
+			"integrity": "sha512-RvwwcruNjI1ncT5xRakeyS9Lf8lcItv34KD+aif+VH9kduAyfYBipGh12274xtenIPZ119/R9BdTBa8gAwSh0A==",
+			"dev": true,
+			"license": "MIT",
+			"engines": {
+				"node": ">=12"
+			},
+			"funding": {
+				"url": "https://github.com/sponsors/jonschlinkert"
+			}
+		},
 		"node_modules/web-streams-polyfill": {
 			"version": "3.3.3",
 			"resolved": "https://registry.npmjs.org/web-streams-polyfill/-/web-streams-polyfill-3.3.3.tgz",
@@ -5101,92 +5471,6 @@
 				}
 			}
 		},
-		"packages/agent/node_modules/@vitest/expect": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/expect/-/expect-4.1.9.tgz",
-			"integrity": "sha512-vl/rYsUKcBr3SnQn166+XR5ZQcgMx3DQhFWdfli/cWpLnLUmbxZvyrJZotLFUryib+LtArYMSTJ5RbQ57ZqrlA==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@standard-schema/spec": "^1.1.0",
-				"@types/chai": "^5.2.2",
-				"@vitest/spy": "4.1.9",
-				"@vitest/utils": "4.1.9",
-				"chai": "^6.2.2",
-				"tinyrainbow": "^3.1.0"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/agent/node_modules/@vitest/pretty-format": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/pretty-format/-/pretty-format-4.1.9.tgz",
-			"integrity": "sha512-s0iufns3iIFitdgm+YR7g1whCAaGtXz459VS9/PqyKDEEFgYIhsHOQmXgIgDuYCt7DeQmiZT0Qe2OA2p4ZPu5A==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"tinyrainbow": "^3.1.0"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/agent/node_modules/@vitest/runner": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/runner/-/runner-4.1.9.tgz",
-			"integrity": "sha512-KXLMDtc7oe70+3mJfGrPUWPesswH+3sTxAMAMl8DG7I8IUQT4XW718dY5ID3vPUcmlu27CcKfY4P3h3I29SLJg==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@vitest/utils": "4.1.9",
-				"pathe": "^2.0.3"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/agent/node_modules/@vitest/snapshot": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/snapshot/-/snapshot-4.1.9.tgz",
-			"integrity": "sha512-Jc7RKGNBo8Z28WYIm0Niej4xdSPByRf6mU58VpHQkd6Zh05rlnA+twjbK5HyeIGHxrzsc3mJgS43uM0CZKzaIA==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@vitest/pretty-format": "4.1.9",
-				"@vitest/utils": "4.1.9",
-				"magic-string": "^0.30.21",
-				"pathe": "^2.0.3"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/agent/node_modules/@vitest/spy": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/spy/-/spy-4.1.9.tgz",
-			"integrity": "sha512-fHpsS6mIi+PiEW+vcRVOMkX1oSaPKne3VOclSFICPcGOmfKgXPU5iAah+wcNcj2xPrCCmfq99IDGf+EojhhvhA==",
-			"dev": true,
-			"license": "MIT",
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/agent/node_modules/@vitest/utils": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/utils/-/utils-4.1.9.tgz",
-			"integrity": "sha512-A51o8ymO5PpqlWNnBP9ZHPXDIpuMtTLlGSjN7la4US+LJzoUMyhwjA5QXlm39JexgwHKW4Xjs8Z2d3dLCXOeuA==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@vitest/pretty-format": "4.1.9",
-				"convert-source-map": "^2.0.0",
-				"tinyrainbow": "^3.1.0"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
 		"packages/agent/node_modules/ast-v8-to-istanbul": {
 			"version": "1.0.4",
 			"resolved": "https://registry.npmjs.org/ast-v8-to-istanbul/-/ast-v8-to-istanbul-1.0.4.tgz",
@@ -5199,23 +5483,6 @@
 				"js-tokens": "^10.0.0"
 			}
 		},
-		"packages/agent/node_modules/chai": {
-			"version": "6.2.2",
-			"resolved": "https://registry.npmjs.org/chai/-/chai-6.2.2.tgz",
-			"integrity": "sha512-NUPRluOfOiTKBKvWPtSD4PhFvWCqOi0BGStNWs57X9js7XGTprSmFoz5F0tWhR4WPjNeR9jXqdC7/UpSJTnlRg==",
-			"dev": true,
-			"license": "MIT",
-			"engines": {
-				"node": ">=18"
-			}
-		},
-		"packages/agent/node_modules/es-module-lexer": {
-			"version": "2.1.0",
-			"resolved": "https://registry.npmjs.org/es-module-lexer/-/es-module-lexer-2.1.0.tgz",
-			"integrity": "sha512-n27zTYMjYu1aj4MjCWzSP7G9r75utsaoc8m61weK+W8JMBGGQybd43GstCXZ3WNmSFtGT9wi59qQTW6mhTR5LQ==",
-			"dev": true,
-			"license": "MIT"
-		},
 		"packages/agent/node_modules/magicast": {
 			"version": "0.5.3",
 			"resolved": "https://registry.npmjs.org/magicast/-/magicast-0.5.3.tgz",
@@ -5228,46 +5495,6 @@
 				"source-map-js": "^1.2.1"
 			}
 		},
-		"packages/agent/node_modules/picomatch": {
-			"version": "4.0.4",
-			"resolved": "https://registry.npmjs.org/picomatch/-/picomatch-4.0.4.tgz",
-			"integrity": "sha512-QP88BAKvMam/3NxH6vj2o21R6MjxZUAd6nlwAS/pnGvN9IVLocLHxGYIzFhg6fUQ+5th6P4dv4eW9jX3DSIj7A==",
-			"dev": true,
-			"license": "MIT",
-			"engines": {
-				"node": ">=12"
-			},
-			"funding": {
-				"url": "https://github.com/sponsors/jonschlinkert"
-			}
-		},
-		"packages/agent/node_modules/std-env": {
-			"version": "4.1.0",
-			"resolved": "https://registry.npmjs.org/std-env/-/std-env-4.1.0.tgz",
-			"integrity": "sha512-Rq7ybcX2RuC55r9oaPVEW7/xu3tj8u4GeBYHBWCychFtzMIr86A7e3PPEBPT37sHStKX3+TiX/Fr/ACmJLVlLQ==",
-			"dev": true,
-			"license": "MIT"
-		},
-		"packages/agent/node_modules/tinyexec": {
-			"version": "1.2.4",
-			"resolved": "https://registry.npmjs.org/tinyexec/-/tinyexec-1.2.4.tgz",
-			"integrity": "sha512-SHf/r48b7vOrjve9PxJo3MN5v5yuyjHvdUcrQffT3WXMUfnGmHDVbC4k3sHJaJTgZCwpUplIaAo5ANtMyp3YHg==",
-			"dev": true,
-			"license": "MIT",
-			"engines": {
-				"node": ">=18"
-			}
-		},
-		"packages/agent/node_modules/tinyrainbow": {
-			"version": "3.1.0",
-			"resolved": "https://registry.npmjs.org/tinyrainbow/-/tinyrainbow-3.1.0.tgz",
-			"integrity": "sha512-Bf+ILmBgretUrdJxzXM0SgXLZ3XfiaUuOj/IKQHuTXip+05Xn+uyEYdVg0kYDipTBcLrCVyUzAPz7QmArb0mmw==",
-			"dev": true,
-			"license": "MIT",
-			"engines": {
-				"node": ">=14.0.0"
-			}
-		},
 		"packages/agent/node_modules/undici-types": {
 			"version": "7.16.0",
 			"resolved": "https://registry.npmjs.org/undici-types/-/undici-types-7.16.0.tgz",
@@ -5275,123 +5502,6 @@
 			"dev": true,
 			"license": "MIT"
 		},
-		"packages/agent/node_modules/vitest": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/vitest/-/vitest-4.1.9.tgz",
-			"integrity": "sha512-nE3/LEyc0z87uHYLZebqCUOaJr2hdtuPp7BQ4BosVFnfltxgAvMG08NyrSGlPpOUWvR27c5flSmYFTNr78L9GQ==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@vitest/expect": "4.1.9",
-				"@vitest/mocker": "4.1.9",
-				"@vitest/pretty-format": "4.1.9",
-				"@vitest/runner": "4.1.9",
-				"@vitest/snapshot": "4.1.9",
-				"@vitest/spy": "4.1.9",
-				"@vitest/utils": "4.1.9",
-				"es-module-lexer": "^2.0.0",
-				"expect-type": "^1.3.0",
-				"magic-string": "^0.30.21",
-				"obug": "^2.1.1",
-				"pathe": "^2.0.3",
-				"picomatch": "^4.0.3",
-				"std-env": "^4.0.0-rc.1",
-				"tinybench": "^2.9.0",
-				"tinyexec": "^1.0.2",
-				"tinyglobby": "^0.2.15",
-				"tinyrainbow": "^3.1.0",
-				"vite": "^6.0.0 || ^7.0.0 || ^8.0.0",
-				"why-is-node-running": "^2.3.0"
-			},
-			"bin": {
-				"vitest": "vitest.mjs"
-			},
-			"engines": {
-				"node": "^20.0.0 || ^22.0.0 || >=24.0.0"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			},
-			"peerDependencies": {
-				"@edge-runtime/vm": "*",
-				"@opentelemetry/api": "^1.9.0",
-				"@types/node": "^20.0.0 || ^22.0.0 || >=24.0.0",
-				"@vitest/browser-playwright": "4.1.9",
-				"@vitest/browser-preview": "4.1.9",
-				"@vitest/browser-webdriverio": "4.1.9",
-				"@vitest/coverage-istanbul": "4.1.9",
-				"@vitest/coverage-v8": "4.1.9",
-				"@vitest/ui": "4.1.9",
-				"happy-dom": "*",
-				"jsdom": "*",
-				"vite": "^6.0.0 || ^7.0.0 || ^8.0.0"
-			},
-			"peerDependenciesMeta": {
-				"@edge-runtime/vm": {
-					"optional": true
-				},
-				"@opentelemetry/api": {
-					"optional": true
-				},
-				"@types/node": {
-					"optional": true
-				},
-				"@vitest/browser-playwright": {
-					"optional": true
-				},
-				"@vitest/browser-preview": {
-					"optional": true
-				},
-				"@vitest/browser-webdriverio": {
-					"optional": true
-				},
-				"@vitest/coverage-istanbul": {
-					"optional": true
-				},
-				"@vitest/coverage-v8": {
-					"optional": true
-				},
-				"@vitest/ui": {
-					"optional": true
-				},
-				"happy-dom": {
-					"optional": true
-				},
-				"jsdom": {
-					"optional": true
-				},
-				"vite": {
-					"optional": false
-				}
-			}
-		},
-		"packages/agent/node_modules/vitest/node_modules/@vitest/mocker": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/mocker/-/mocker-4.1.9.tgz",
-			"integrity": "sha512-EVkXzBjrPGM+cK8/ANWgBrkUCfJfb38/EfTSO8h7pWvKkyPkpWxvR7BkD2MyItMF62C97zAEoqdpUixwR/e+Rw==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@vitest/spy": "4.1.9",
-				"estree-walker": "^3.0.3",
-				"magic-string": "^0.30.21"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			},
-			"peerDependencies": {
-				"msw": "^2.4.9",
-				"vite": "^6.0.0 || ^7.0.0 || ^8.0.0"
-			},
-			"peerDependenciesMeta": {
-				"msw": {
-					"optional": true
-				},
-				"vite": {
-					"optional": true
-				}
-			}
-		},
 		"packages/ai": {
 			"name": "@earendil-works/pi-ai",
 			"version": "0.82.0",
@@ -5431,176 +5541,6 @@
 				"undici-types": "~7.16.0"
 			}
 		},
-		"packages/ai/node_modules/@vitest/expect": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/expect/-/expect-4.1.9.tgz",
-			"integrity": "sha512-vl/rYsUKcBr3SnQn166+XR5ZQcgMx3DQhFWdfli/cWpLnLUmbxZvyrJZotLFUryib+LtArYMSTJ5RbQ57ZqrlA==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@standard-schema/spec": "^1.1.0",
-				"@types/chai": "^5.2.2",
-				"@vitest/spy": "4.1.9",
-				"@vitest/utils": "4.1.9",
-				"chai": "^6.2.2",
-				"tinyrainbow": "^3.1.0"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/ai/node_modules/@vitest/mocker": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/mocker/-/mocker-4.1.9.tgz",
-			"integrity": "sha512-EVkXzBjrPGM+cK8/ANWgBrkUCfJfb38/EfTSO8h7pWvKkyPkpWxvR7BkD2MyItMF62C97zAEoqdpUixwR/e+Rw==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@vitest/spy": "4.1.9",
-				"estree-walker": "^3.0.3",
-				"magic-string": "^0.30.21"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			},
-			"peerDependencies": {
-				"msw": "^2.4.9",
-				"vite": "^6.0.0 || ^7.0.0 || ^8.0.0"
-			},
-			"peerDependenciesMeta": {
-				"msw": {
-					"optional": true
-				},
-				"vite": {
-					"optional": true
-				}
-			}
-		},
-		"packages/ai/node_modules/@vitest/pretty-format": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/pretty-format/-/pretty-format-4.1.9.tgz",
-			"integrity": "sha512-s0iufns3iIFitdgm+YR7g1whCAaGtXz459VS9/PqyKDEEFgYIhsHOQmXgIgDuYCt7DeQmiZT0Qe2OA2p4ZPu5A==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"tinyrainbow": "^3.1.0"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/ai/node_modules/@vitest/runner": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/runner/-/runner-4.1.9.tgz",
-			"integrity": "sha512-KXLMDtc7oe70+3mJfGrPUWPesswH+3sTxAMAMl8DG7I8IUQT4XW718dY5ID3vPUcmlu27CcKfY4P3h3I29SLJg==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@vitest/utils": "4.1.9",
-				"pathe": "^2.0.3"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/ai/node_modules/@vitest/snapshot": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/snapshot/-/snapshot-4.1.9.tgz",
-			"integrity": "sha512-Jc7RKGNBo8Z28WYIm0Niej4xdSPByRf6mU58VpHQkd6Zh05rlnA+twjbK5HyeIGHxrzsc3mJgS43uM0CZKzaIA==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@vitest/pretty-format": "4.1.9",
-				"@vitest/utils": "4.1.9",
-				"magic-string": "^0.30.21",
-				"pathe": "^2.0.3"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/ai/node_modules/@vitest/spy": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/spy/-/spy-4.1.9.tgz",
-			"integrity": "sha512-fHpsS6mIi+PiEW+vcRVOMkX1oSaPKne3VOclSFICPcGOmfKgXPU5iAah+wcNcj2xPrCCmfq99IDGf+EojhhvhA==",
-			"dev": true,
-			"license": "MIT",
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/ai/node_modules/@vitest/utils": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/utils/-/utils-4.1.9.tgz",
-			"integrity": "sha512-A51o8ymO5PpqlWNnBP9ZHPXDIpuMtTLlGSjN7la4US+LJzoUMyhwjA5QXlm39JexgwHKW4Xjs8Z2d3dLCXOeuA==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@vitest/pretty-format": "4.1.9",
-				"convert-source-map": "^2.0.0",
-				"tinyrainbow": "^3.1.0"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/ai/node_modules/chai": {
-			"version": "6.2.2",
-			"resolved": "https://registry.npmjs.org/chai/-/chai-6.2.2.tgz",
-			"integrity": "sha512-NUPRluOfOiTKBKvWPtSD4PhFvWCqOi0BGStNWs57X9js7XGTprSmFoz5F0tWhR4WPjNeR9jXqdC7/UpSJTnlRg==",
-			"dev": true,
-			"license": "MIT",
-			"engines": {
-				"node": ">=18"
-			}
-		},
-		"packages/ai/node_modules/es-module-lexer": {
-			"version": "2.1.0",
-			"resolved": "https://registry.npmjs.org/es-module-lexer/-/es-module-lexer-2.1.0.tgz",
-			"integrity": "sha512-n27zTYMjYu1aj4MjCWzSP7G9r75utsaoc8m61weK+W8JMBGGQybd43GstCXZ3WNmSFtGT9wi59qQTW6mhTR5LQ==",
-			"dev": true,
-			"license": "MIT"
-		},
-		"packages/ai/node_modules/picomatch": {
-			"version": "4.0.4",
-			"resolved": "https://registry.npmjs.org/picomatch/-/picomatch-4.0.4.tgz",
-			"integrity": "sha512-QP88BAKvMam/3NxH6vj2o21R6MjxZUAd6nlwAS/pnGvN9IVLocLHxGYIzFhg6fUQ+5th6P4dv4eW9jX3DSIj7A==",
-			"dev": true,
-			"license": "MIT",
-			"engines": {
-				"node": ">=12"
-			},
-			"funding": {
-				"url": "https://github.com/sponsors/jonschlinkert"
-			}
-		},
-		"packages/ai/node_modules/std-env": {
-			"version": "4.1.0",
-			"resolved": "https://registry.npmjs.org/std-env/-/std-env-4.1.0.tgz",
-			"integrity": "sha512-Rq7ybcX2RuC55r9oaPVEW7/xu3tj8u4GeBYHBWCychFtzMIr86A7e3PPEBPT37sHStKX3+TiX/Fr/ACmJLVlLQ==",
-			"dev": true,
-			"license": "MIT"
-		},
-		"packages/ai/node_modules/tinyexec": {
-			"version": "1.2.4",
-			"resolved": "https://registry.npmjs.org/tinyexec/-/tinyexec-1.2.4.tgz",
-			"integrity": "sha512-SHf/r48b7vOrjve9PxJo3MN5v5yuyjHvdUcrQffT3WXMUfnGmHDVbC4k3sHJaJTgZCwpUplIaAo5ANtMyp3YHg==",
-			"dev": true,
-			"license": "MIT",
-			"engines": {
-				"node": ">=18"
-			}
-		},
-		"packages/ai/node_modules/tinyrainbow": {
-			"version": "3.1.0",
-			"resolved": "https://registry.npmjs.org/tinyrainbow/-/tinyrainbow-3.1.0.tgz",
-			"integrity": "sha512-Bf+ILmBgretUrdJxzXM0SgXLZ3XfiaUuOj/IKQHuTXip+05Xn+uyEYdVg0kYDipTBcLrCVyUzAPz7QmArb0mmw==",
-			"dev": true,
-			"license": "MIT",
-			"engines": {
-				"node": ">=14.0.0"
-			}
-		},
 		"packages/ai/node_modules/undici-types": {
 			"version": "7.16.0",
 			"resolved": "https://registry.npmjs.org/undici-types/-/undici-types-7.16.0.tgz",
@@ -5608,96 +5548,6 @@
 			"dev": true,
 			"license": "MIT"
 		},
-		"packages/ai/node_modules/vitest": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/vitest/-/vitest-4.1.9.tgz",
-			"integrity": "sha512-nE3/LEyc0z87uHYLZebqCUOaJr2hdtuPp7BQ4BosVFnfltxgAvMG08NyrSGlPpOUWvR27c5flSmYFTNr78L9GQ==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@vitest/expect": "4.1.9",
-				"@vitest/mocker": "4.1.9",
-				"@vitest/pretty-format": "4.1.9",
-				"@vitest/runner": "4.1.9",
-				"@vitest/snapshot": "4.1.9",
-				"@vitest/spy": "4.1.9",
-				"@vitest/utils": "4.1.9",
-				"es-module-lexer": "^2.0.0",
-				"expect-type": "^1.3.0",
-				"magic-string": "^0.30.21",
-				"obug": "^2.1.1",
-				"pathe": "^2.0.3",
-				"picomatch": "^4.0.3",
-				"std-env": "^4.0.0-rc.1",
-				"tinybench": "^2.9.0",
-				"tinyexec": "^1.0.2",
-				"tinyglobby": "^0.2.15",
-				"tinyrainbow": "^3.1.0",
-				"vite": "^6.0.0 || ^7.0.0 || ^8.0.0",
-				"why-is-node-running": "^2.3.0"
-			},
-			"bin": {
-				"vitest": "vitest.mjs"
-			},
-			"engines": {
-				"node": "^20.0.0 || ^22.0.0 || >=24.0.0"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			},
-			"peerDependencies": {
-				"@edge-runtime/vm": "*",
-				"@opentelemetry/api": "^1.9.0",
-				"@types/node": "^20.0.0 || ^22.0.0 || >=24.0.0",
-				"@vitest/browser-playwright": "4.1.9",
-				"@vitest/browser-preview": "4.1.9",
-				"@vitest/browser-webdriverio": "4.1.9",
-				"@vitest/coverage-istanbul": "4.1.9",
-				"@vitest/coverage-v8": "4.1.9",
-				"@vitest/ui": "4.1.9",
-				"happy-dom": "*",
-				"jsdom": "*",
-				"vite": "^6.0.0 || ^7.0.0 || ^8.0.0"
-			},
-			"peerDependenciesMeta": {
-				"@edge-runtime/vm": {
-					"optional": true
-				},
-				"@opentelemetry/api": {
-					"optional": true
-				},
-				"@types/node": {
-					"optional": true
-				},
-				"@vitest/browser-playwright": {
-					"optional": true
-				},
-				"@vitest/browser-preview": {
-					"optional": true
-				},
-				"@vitest/browser-webdriverio": {
-					"optional": true
-				},
-				"@vitest/coverage-istanbul": {
-					"optional": true
-				},
-				"@vitest/coverage-v8": {
-					"optional": true
-				},
-				"@vitest/ui": {
-					"optional": true
-				},
-				"happy-dom": {
-					"optional": true
-				},
-				"jsdom": {
-					"optional": true
-				},
-				"vite": {
-					"optional": false
-				}
-			}
-		},
 		"packages/coding-agent": {
 			"name": "@earendil-works/pi-coding-agent",
 			"version": "0.82.0",
@@ -5819,149 +5669,6 @@
 				"undici-types": "~7.16.0"
 			}
 		},
-		"packages/coding-agent/node_modules/@vitest/expect": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/expect/-/expect-4.1.9.tgz",
-			"integrity": "sha512-vl/rYsUKcBr3SnQn166+XR5ZQcgMx3DQhFWdfli/cWpLnLUmbxZvyrJZotLFUryib+LtArYMSTJ5RbQ57ZqrlA==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@standard-schema/spec": "^1.1.0",
-				"@types/chai": "^5.2.2",
-				"@vitest/spy": "4.1.9",
-				"@vitest/utils": "4.1.9",
-				"chai": "^6.2.2",
-				"tinyrainbow": "^3.1.0"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/coding-agent/node_modules/@vitest/pretty-format": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/pretty-format/-/pretty-format-4.1.9.tgz",
-			"integrity": "sha512-s0iufns3iIFitdgm+YR7g1whCAaGtXz459VS9/PqyKDEEFgYIhsHOQmXgIgDuYCt7DeQmiZT0Qe2OA2p4ZPu5A==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"tinyrainbow": "^3.1.0"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/coding-agent/node_modules/@vitest/runner": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/runner/-/runner-4.1.9.tgz",
-			"integrity": "sha512-KXLMDtc7oe70+3mJfGrPUWPesswH+3sTxAMAMl8DG7I8IUQT4XW718dY5ID3vPUcmlu27CcKfY4P3h3I29SLJg==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@vitest/utils": "4.1.9",
-				"pathe": "^2.0.3"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/coding-agent/node_modules/@vitest/snapshot": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/snapshot/-/snapshot-4.1.9.tgz",
-			"integrity": "sha512-Jc7RKGNBo8Z28WYIm0Niej4xdSPByRf6mU58VpHQkd6Zh05rlnA+twjbK5HyeIGHxrzsc3mJgS43uM0CZKzaIA==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@vitest/pretty-format": "4.1.9",
-				"@vitest/utils": "4.1.9",
-				"magic-string": "^0.30.21",
-				"pathe": "^2.0.3"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/coding-agent/node_modules/@vitest/spy": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/spy/-/spy-4.1.9.tgz",
-			"integrity": "sha512-fHpsS6mIi+PiEW+vcRVOMkX1oSaPKne3VOclSFICPcGOmfKgXPU5iAah+wcNcj2xPrCCmfq99IDGf+EojhhvhA==",
-			"dev": true,
-			"license": "MIT",
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/coding-agent/node_modules/@vitest/utils": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/utils/-/utils-4.1.9.tgz",
-			"integrity": "sha512-A51o8ymO5PpqlWNnBP9ZHPXDIpuMtTLlGSjN7la4US+LJzoUMyhwjA5QXlm39JexgwHKW4Xjs8Z2d3dLCXOeuA==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@vitest/pretty-format": "4.1.9",
-				"convert-source-map": "^2.0.0",
-				"tinyrainbow": "^3.1.0"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			}
-		},
-		"packages/coding-agent/node_modules/chai": {
-			"version": "6.2.2",
-			"resolved": "https://registry.npmjs.org/chai/-/chai-6.2.2.tgz",
-			"integrity": "sha512-NUPRluOfOiTKBKvWPtSD4PhFvWCqOi0BGStNWs57X9js7XGTprSmFoz5F0tWhR4WPjNeR9jXqdC7/UpSJTnlRg==",
-			"dev": true,
-			"license": "MIT",
-			"engines": {
-				"node": ">=18"
-			}
-		},
-		"packages/coding-agent/node_modules/es-module-lexer": {
-			"version": "2.1.0",
-			"resolved": "https://registry.npmjs.org/es-module-lexer/-/es-module-lexer-2.1.0.tgz",
-			"integrity": "sha512-n27zTYMjYu1aj4MjCWzSP7G9r75utsaoc8m61weK+W8JMBGGQybd43GstCXZ3WNmSFtGT9wi59qQTW6mhTR5LQ==",
-			"dev": true,
-			"license": "MIT"
-		},
-		"packages/coding-agent/node_modules/picomatch": {
-			"version": "4.0.4",
-			"resolved": "https://registry.npmjs.org/picomatch/-/picomatch-4.0.4.tgz",
-			"integrity": "sha512-QP88BAKvMam/3NxH6vj2o21R6MjxZUAd6nlwAS/pnGvN9IVLocLHxGYIzFhg6fUQ+5th6P4dv4eW9jX3DSIj7A==",
-			"dev": true,
-			"license": "MIT",
-			"engines": {
-				"node": ">=12"
-			},
-			"funding": {
-				"url": "https://github.com/sponsors/jonschlinkert"
-			}
-		},
-		"packages/coding-agent/node_modules/std-env": {
-			"version": "4.1.0",
-			"resolved": "https://registry.npmjs.org/std-env/-/std-env-4.1.0.tgz",
-			"integrity": "sha512-Rq7ybcX2RuC55r9oaPVEW7/xu3tj8u4GeBYHBWCychFtzMIr86A7e3PPEBPT37sHStKX3+TiX/Fr/ACmJLVlLQ==",
-			"dev": true,
-			"license": "MIT"
-		},
-		"packages/coding-agent/node_modules/tinyexec": {
-			"version": "1.2.4",
-			"resolved": "https://registry.npmjs.org/tinyexec/-/tinyexec-1.2.4.tgz",
-			"integrity": "sha512-SHf/r48b7vOrjve9PxJo3MN5v5yuyjHvdUcrQffT3WXMUfnGmHDVbC4k3sHJaJTgZCwpUplIaAo5ANtMyp3YHg==",
-			"dev": true,
-			"license": "MIT",
-			"engines": {
-				"node": ">=18"
-			}
-		},
-		"packages/coding-agent/node_modules/tinyrainbow": {
-			"version": "3.1.0",
-			"resolved": "https://registry.npmjs.org/tinyrainbow/-/tinyrainbow-3.1.0.tgz",
-			"integrity": "sha512-Bf+ILmBgretUrdJxzXM0SgXLZ3XfiaUuOj/IKQHuTXip+05Xn+uyEYdVg0kYDipTBcLrCVyUzAPz7QmArb0mmw==",
-			"dev": true,
-			"license": "MIT",
-			"engines": {
-				"node": ">=14.0.0"
-			}
-		},
 		"packages/coding-agent/node_modules/undici-types": {
 			"version": "7.16.0",
 			"resolved": "https://registry.npmjs.org/undici-types/-/undici-types-7.16.0.tgz",
@@ -5969,123 +5676,83 @@
 			"dev": true,
 			"license": "MIT"
 		},
-		"packages/coding-agent/node_modules/vitest": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/vitest/-/vitest-4.1.9.tgz",
-			"integrity": "sha512-nE3/LEyc0z87uHYLZebqCUOaJr2hdtuPp7BQ4BosVFnfltxgAvMG08NyrSGlPpOUWvR27c5flSmYFTNr78L9GQ==",
-			"dev": true,
-			"license": "MIT",
-			"dependencies": {
-				"@vitest/expect": "4.1.9",
-				"@vitest/mocker": "4.1.9",
-				"@vitest/pretty-format": "4.1.9",
-				"@vitest/runner": "4.1.9",
-				"@vitest/snapshot": "4.1.9",
-				"@vitest/spy": "4.1.9",
-				"@vitest/utils": "4.1.9",
-				"es-module-lexer": "^2.0.0",
-				"expect-type": "^1.3.0",
-				"magic-string": "^0.30.21",
-				"obug": "^2.1.1",
-				"pathe": "^2.0.3",
-				"picomatch": "^4.0.3",
-				"std-env": "^4.0.0-rc.1",
-				"tinybench": "^2.9.0",
-				"tinyexec": "^1.0.2",
-				"tinyglobby": "^0.2.15",
-				"tinyrainbow": "^3.1.0",
-				"vite": "^6.0.0 || ^7.0.0 || ^8.0.0",
-				"why-is-node-running": "^2.3.0"
-			},
-			"bin": {
-				"vitest": "vitest.mjs"
-			},
-			"engines": {
-				"node": "^20.0.0 || ^22.0.0 || >=24.0.0"
-			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			},
-			"peerDependencies": {
-				"@edge-runtime/vm": "*",
-				"@opentelemetry/api": "^1.9.0",
-				"@types/node": "^20.0.0 || ^22.0.0 || >=24.0.0",
-				"@vitest/browser-playwright": "4.1.9",
-				"@vitest/browser-preview": "4.1.9",
-				"@vitest/browser-webdriverio": "4.1.9",
-				"@vitest/coverage-istanbul": "4.1.9",
-				"@vitest/coverage-v8": "4.1.9",
-				"@vitest/ui": "4.1.9",
-				"happy-dom": "*",
-				"jsdom": "*",
-				"vite": "^6.0.0 || ^7.0.0 || ^8.0.0"
-			},
-			"peerDependenciesMeta": {
-				"@edge-runtime/vm": {
-					"optional": true
-				},
-				"@opentelemetry/api": {
-					"optional": true
-				},
-				"@types/node": {
-					"optional": true
-				},
-				"@vitest/browser-playwright": {
-					"optional": true
-				},
-				"@vitest/browser-preview": {
-					"optional": true
-				},
-				"@vitest/browser-webdriverio": {
-					"optional": true
-				},
-				"@vitest/coverage-istanbul": {
-					"optional": true
-				},
-				"@vitest/coverage-v8": {
-					"optional": true
-				},
-				"@vitest/ui": {
-					"optional": true
-				},
-				"happy-dom": {
-					"optional": true
-				},
-				"jsdom": {
-					"optional": true
-				},
-				"vite": {
-					"optional": false
-				}
+		"packages/evals": {
+			"name": "@earendil-works/pi-evals",
+			"version": "0.82.0",
+			"devDependencies": {
+				"@earendil-works/pi-coding-agent": "^0.82.0",
+				"@mariozechner/pi-agent-core": "npm:@earendil-works/pi-agent-core@0.82.0",
+				"@mariozechner/pi-ai": "npm:@earendil-works/pi-ai@0.82.0",
+				"@types/node": "24.12.4",
+				"@vitest-evals/harness-pi-ai": "0.15.0",
+				"shx": "0.4.0",
+				"typescript": "5.9.3",
+				"vitest": "4.1.9",
+				"vitest-evals": "0.15.0"
 			}
 		},
-		"packages/coding-agent/node_modules/vitest/node_modules/@vitest/mocker": {
-			"version": "4.1.9",
-			"resolved": "https://registry.npmjs.org/@vitest/mocker/-/mocker-4.1.9.tgz",
-			"integrity": "sha512-EVkXzBjrPGM+cK8/ANWgBrkUCfJfb38/EfTSO8h7pWvKkyPkpWxvR7BkD2MyItMF62C97zAEoqdpUixwR/e+Rw==",
+		"packages/evals/node_modules/@mariozechner/pi-agent-core": {
+			"name": "@earendil-works/pi-agent-core",
+			"version": "0.82.0",
+			"resolved": "https://registry.npmjs.org/@earendil-works/pi-agent-core/-/pi-agent-core-0.82.0.tgz",
+			"integrity": "sha512-bnS9DpOKK5T/F/gQkaOnYdMsuuciWiScfAHHWC+k5OQ0HxjSqMFQvp8keurULLoT4+v8NHv4V14pNvd4hsfC0Q==",
 			"dev": true,
 			"license": "MIT",
 			"dependencies": {
-				"@vitest/spy": "4.1.9",
-				"estree-walker": "^3.0.3",
-				"magic-string": "^0.30.21"
+				"@earendil-works/pi-ai": "^0.82.0",
+				"diff": "8.0.4",
+				"ignore": "7.0.5",
+				"typebox": "1.1.38",
+				"yaml": "2.9.0"
 			},
-			"funding": {
-				"url": "https://opencollective.com/vitest"
-			},
-			"peerDependencies": {
-				"msw": "^2.4.9",
-				"vite": "^6.0.0 || ^7.0.0 || ^8.0.0"
-			},
-			"peerDependenciesMeta": {
-				"msw": {
-					"optional": true
-				},
-				"vite": {
-					"optional": true
-				}
+			"engines": {
+				"node": ">=22.19.0"
 			}
 		},
+		"packages/evals/node_modules/@mariozechner/pi-ai": {
+			"name": "@earendil-works/pi-ai",
+			"version": "0.82.0",
+			"resolved": "https://registry.npmjs.org/@earendil-works/pi-ai/-/pi-ai-0.82.0.tgz",
+			"integrity": "sha512-8MvW9+zno13sXDuT2kFMnWeTNUufUhPeZDRVO+igGoBRCDWgn7Xh2FkRQI1mRuet6QhF4ENQuLYdIAOyG6BhNw==",
+			"dev": true,
+			"license": "MIT",
+			"dependencies": {
+				"@anthropic-ai/sdk": "0.91.1",
+				"@aws-sdk/client-bedrock-runtime": "3.1048.0",
+				"@google/genai": "1.52.0",
+				"@mistralai/mistralai": "2.2.6",
+				"@opentelemetry/api": "1.9.0",
+				"@smithy/node-http-handler": "4.7.3",
+				"http-proxy-agent": "7.0.2",
+				"https-proxy-agent": "7.0.6",
+				"openai": "6.26.0",
+				"partial-json": "0.1.7",
+				"typebox": "1.1.38"
+			},
+			"bin": {
+				"pi-ai": "dist/cli.js"
+			},
+			"engines": {
+				"node": ">=22.19.0"
+			}
+		},
+		"packages/evals/node_modules/@types/node": {
+			"version": "24.12.4",
+			"resolved": "https://registry.npmjs.org/@types/node/-/node-24.12.4.tgz",
+			"integrity": "sha512-GUUEShf+PBCGW2KaXwcIt3Yk+e3pkKwWKb9GSyM9WQVE+ep2jzmHdGsHzu4wgcZy5fN9FBdVzjpBQsYlpfpgLA==",
+			"dev": true,
+			"license": "MIT",
+			"dependencies": {
+				"undici-types": "~7.16.0"
+			}
+		},
+		"packages/evals/node_modules/undici-types": {
+			"version": "7.16.0",
+			"resolved": "https://registry.npmjs.org/undici-types/-/undici-types-7.16.0.tgz",
+			"integrity": "sha512-Zz+aZWSj8LE6zoxD+xrjh4VfkIG8Ya6LvYkZqtUQGJPZjYl53ypCaUwWqo7eI0x66KBGeRo+mlBEkMSeSZ38Nw==",
+			"dev": true,
+			"license": "MIT"
+		},
 		"packages/server": {
 			"name": "@earendil-works/pi-server",
 			"version": "0.82.0",
diff --git a/package.json b/package.json
index ae0bb74fa..74979fc70 100644
--- a/package.json
+++ b/package.json
@@ -26,10 +26,12 @@
 		"check:model-data": "npm --prefix packages/ai run check:model-data",
 		"generate:model-catalog": "npm --prefix packages/ai run generate-model-catalog",
 		"diff:model-catalog": "node scripts/diff-model-catalog.mjs",
+		"eval": "npm run eval --workspace=@earendil-works/pi-evals --",
 		"check:model-catalog": "node scripts/publish-model-catalog.mjs --input .artifacts/model-catalog --dry-run",
 		"profile:tui": "node scripts/profile-coding-agent-node.mjs --mode tui",
 		"profile:rpc": "node scripts/profile-coding-agent-node.mjs --mode rpc",
-		"test": "npm run test --workspaces --if-present",
+		"test": "npm run test:scripts && npm run test --workspaces --if-present",
+		"test:scripts": "node --test scripts/*.test.mjs",
 		"version:patch": "npm version patch -ws --no-git-tag-version && node scripts/sync-versions.js && npm install --package-lock-only --ignore-scripts",
 		"version:minor": "npm version minor -ws --no-git-tag-version && node scripts/sync-versions.js && npm install --package-lock-only --ignore-scripts",
 		"version:major": "npm version major -ws --no-git-tag-version && node scripts/sync-versions.js && npm install --package-lock-only --ignore-scripts",
diff --git a/packages/coding-agent/vitest.config.ts b/packages/coding-agent/vitest.config.ts
index 2c6b8cf33..dbd8f34ac 100644
--- a/packages/coding-agent/vitest.config.ts
+++ b/packages/coding-agent/vitest.config.ts
@@ -1,38 +1,28 @@
-import { fileURLToPath } from "node:url";
-import { defineConfig } from "vitest/config";
+import { defineConfig, mergeConfig } from "vitest/config";
+import baseConfig, { workspaceSourcePaths } from "../../vitest.base.ts";
 
-const aiSrcIndex = fileURLToPath(new URL("../ai/src/index.ts", import.meta.url));
-const aiSrcCompat = fileURLToPath(new URL("../ai/src/compat.ts", import.meta.url));
-const aiSrcOAuth = fileURLToPath(new URL("../ai/src/oauth.ts", import.meta.url));
-const aiSrcProviders = fileURLToPath(new URL("../ai/src/providers", import.meta.url));
-const agentSrcIndex = fileURLToPath(new URL("../agent/src/index.ts", import.meta.url));
-const tuiSrcIndex = fileURLToPath(new URL("../tui/src/index.ts", import.meta.url));
-
-export default defineConfig({
-	test: {
-		globals: true,
-		environment: "node",
-		testTimeout: 30000,
-		reporters: process.env.GITHUB_ACTIONS ? ["dot", "github-actions"] : ["dot"],
-		silent: "passed-only",
-		server: {
-			deps: {
-				external: [/@silvia-odwyer\/photon-node/],
+export default mergeConfig(
+	baseConfig,
+	defineConfig({
+		test: {
+			globals: true,
+			environment: "node",
+			testTimeout: 30000,
+			reporters: process.env.GITHUB_ACTIONS ? ["dot", "github-actions"] : ["dot"],
+			silent: "passed-only",
+			server: {
+				deps: {
+					external: [/@silvia-odwyer\/photon-node/],
+				},
 			},
 		},
-	},
-	resolve: {
-		alias: [
-			{ find: /^@earendil-works\/pi-ai$/, replacement: aiSrcIndex },
-			{ find: /^@earendil-works\/pi-ai\/compat$/, replacement: aiSrcCompat },
-			{ find: /^@earendil-works\/pi-ai\/oauth$/, replacement: aiSrcOAuth },
-			{ find: /^@earendil-works\/pi-ai\/providers\/(.+)$/, replacement: `${aiSrcProviders}/$1.ts` },
-			{ find: /^@earendil-works\/pi-agent-core$/, replacement: agentSrcIndex },
-			{ find: /^@earendil-works\/pi-tui$/, replacement: tuiSrcIndex },
-			{ find: /^@mariozechner\/pi-ai$/, replacement: aiSrcIndex },
-			{ find: /^@mariozechner\/pi-ai\/oauth$/, replacement: aiSrcOAuth },
-			{ find: /^@mariozechner\/pi-agent-core$/, replacement: agentSrcIndex },
-			{ find: /^@mariozechner\/pi-tui$/, replacement: tuiSrcIndex },
-		],
-	},
-});
+		resolve: {
+			alias: [
+				{ find: /^@mariozechner\/pi-ai$/, replacement: workspaceSourcePaths.aiIndex },
+				{ find: /^@mariozechner\/pi-ai\/oauth$/, replacement: workspaceSourcePaths.aiOAuth },
+				{ find: /^@mariozechner\/pi-agent-core$/, replacement: workspaceSourcePaths.agentIndex },
+				{ find: /^@mariozechner\/pi-tui$/, replacement: workspaceSourcePaths.tuiIndex },
+			],
+		},
+	}),
+);
diff --git a/packages/evals/README.md b/packages/evals/README.md
new file mode 100644
index 000000000..2e938cb0a
--- /dev/null
+++ b/packages/evals/README.md
@@ -0,0 +1,34 @@
+# Pi evals
+
+Behavioral evaluations for Pi using `vitest-evals` and its Pi harness.
+
+## Running
+
+From the repository root, run with an explicit provider and model:
+
+```bash
+npm run eval -- --provider openai-codex --model gpt-5.4
+```
+
+When invoked from a Pi Bash tool, the current session supplies `PI_PROVIDER` and `PI_MODEL`, so this is sufficient:
+
+```bash
+npm run eval
+```
+
+The runner requires both values and never falls back to another model. Additional arguments are forwarded to Vitest, for example:
+
+```bash
+npm run eval -- -t "capital of France"
+```
+
+Authentication is resolved by Pi's normal `ModelRuntime`. Subscription-backed providers such as `openai-codex` use credentials from the user's Pi configuration. API-backed providers use their standard environment variables, such as `OPENAI_API_KEY` for `openai`.
+
+## Pi harness compatibility
+
+`@vitest-evals/harness-pi-ai` 0.15.0 still declares and imports the former
+`@mariozechner/pi-ai` and `@mariozechner/pi-agent-core` package names. The eval
+package provides aliases at the current Pi version and routes the legacy AI
+import through `src/pi-ai-compat.ts`. Keep those alias versions synchronized
+with the Pi workspace version until the upstream harness supports the current
+`@earendil-works` packages.
diff --git a/packages/evals/package.json b/packages/evals/package.json
new file mode 100644
index 000000000..ab7773ffb
--- /dev/null
+++ b/packages/evals/package.json
@@ -0,0 +1,21 @@
+{
+	"name": "@earendil-works/pi-evals",
+	"version": "0.82.0",
+	"private": true,
+	"type": "module",
+	"scripts": {
+		"clean": "shx rm -rf .eval",
+		"eval": "node scripts/run-evals.mjs"
+	},
+	"devDependencies": {
+		"@earendil-works/pi-coding-agent": "^0.82.0",
+		"@mariozechner/pi-agent-core": "npm:@earendil-works/pi-agent-core@0.82.0",
+		"@mariozechner/pi-ai": "npm:@earendil-works/pi-ai@0.82.0",
+		"@types/node": "24.12.4",
+		"@vitest-evals/harness-pi-ai": "0.15.0",
+		"shx": "0.4.0",
+		"typescript": "5.9.3",
+		"vitest-evals": "0.15.0",
+		"vitest": "4.1.9"
+	}
+}
diff --git a/packages/evals/scripts/run-evals.mjs b/packages/evals/scripts/run-evals.mjs
new file mode 100644
index 000000000..84799e09d
--- /dev/null
+++ b/packages/evals/scripts/run-evals.mjs
@@ -0,0 +1,78 @@
+import { spawnSync } from "node:child_process";
+import { createRequire } from "node:module";
+import { dirname, resolve } from "node:path";
+import { fileURLToPath } from "node:url";
+
+const packageRoot = resolve(dirname(fileURLToPath(import.meta.url)), "..");
+const args = process.argv.slice(2);
+let provider;
+let model;
+let hasCliModelSelection = false;
+const vitestArgs = [];
+
+for (let index = 0; index < args.length; index += 1) {
+	const arg = args[index];
+	if (arg === "--provider" || arg === "--model") {
+		const value = args[index + 1];
+		if (!value || value.startsWith("-")) {
+			console.error(`Missing value for ${arg}`);
+			process.exit(1);
+		}
+		if (arg === "--provider") provider = value;
+		else model = value;
+		hasCliModelSelection = true;
+		index += 1;
+		continue;
+	}
+	if (arg.startsWith("--provider=")) {
+		provider = arg.slice("--provider=".length);
+		hasCliModelSelection = true;
+		continue;
+	}
+	if (arg.startsWith("--model=")) {
+		model = arg.slice("--model=".length);
+		hasCliModelSelection = true;
+		continue;
+	}
+	vitestArgs.push(arg);
+}
+
+if (hasCliModelSelection && (!provider || !model)) {
+	console.error("CLI model selection requires both --provider and --model.");
+	process.exit(1);
+}
+
+provider ??= process.env.PI_PROVIDER;
+model ??= process.env.PI_MODEL;
+
+if (!provider || !model) {
+	console.error(
+		"No eval model selected. Pass --provider and --model, or set PI_PROVIDER and PI_MODEL.",
+	);
+	process.exit(1);
+}
+
+const require = createRequire(import.meta.url);
+const vitestPackagePath = require.resolve("vitest/package.json");
+const vitestCliPath = resolve(dirname(vitestPackagePath), "vitest.mjs");
+
+console.error(`[eval] provider=${provider} model=${model}`);
+const result = spawnSync(
+	process.execPath,
+	[vitestCliPath, "run", "--config", "vitest.config.ts", ...vitestArgs],
+	{
+		cwd: packageRoot,
+		stdio: "inherit",
+		env: {
+			...process.env,
+			PI_PROVIDER: provider,
+			PI_MODEL: model,
+		},
+	},
+);
+
+if (result.error) {
+	throw result.error;
+}
+
+process.exit(result.status ?? 1);
diff --git a/packages/evals/src/general-knowledge.eval.ts b/packages/evals/src/general-knowledge.eval.ts
new file mode 100644
index 000000000..7354bb4ec
--- /dev/null
+++ b/packages/evals/src/general-knowledge.eval.ts
@@ -0,0 +1,15 @@
+import { expect } from "vitest";
+import { describeEval } from "vitest-evals";
+import { piCodingAgentHarness } from "./pi-harness.ts";
+
+describeEval("general knowledge", { harness: piCodingAgentHarness }, (it) => {
+	it("knows the capital of France", async ({ run }) => {
+		const result = await run("What's the capital of France? Respond with only the city name.");
+
+		expect(result.output.trim()).toBe("Paris");
+		expect(result.errors).toEqual([]);
+		expect(result.usage.provider).toBe(process.env.PI_PROVIDER);
+		expect(result.usage.model).toBe(process.env.PI_MODEL);
+		expect(result.usage.totalTokens).toBeGreaterThan(0);
+	});
+});
diff --git a/packages/evals/src/pi-ai-compat.ts b/packages/evals/src/pi-ai-compat.ts
new file mode 100644
index 000000000..7e8f7ed31
--- /dev/null
+++ b/packages/evals/src/pi-ai-compat.ts
@@ -0,0 +1,5 @@
+// @vitest-evals/harness-pi-ai 0.15.0 imports completeSimple from the legacy
+// @mariozechner/pi-ai entry point. Remove this shim when the harness supports
+// the current @earendil-works/pi-ai API.
+export * from "@earendil-works/pi-ai";
+export { completeSimple } from "@earendil-works/pi-ai/compat";
diff --git a/packages/evals/src/pi-harness.ts b/packages/evals/src/pi-harness.ts
new file mode 100644
index 000000000..0585a13ea
--- /dev/null
+++ b/packages/evals/src/pi-harness.ts
@@ -0,0 +1,182 @@
+import { mkdir, mkdtemp, rm } from "node:fs/promises";
+import { tmpdir } from "node:os";
+import { join } from "node:path";
+import {
+	createAgentSession,
+	DefaultResourceLoader,
+	ModelRuntime,
+	SessionManager,
+	SettingsManager,
+} from "@earendil-works/pi-coding-agent";
+import { type PiAiCreateAgentArgs, type PiAiRuntime, type PiAiToolset, piAiHarness } from "@vitest-evals/harness-pi-ai";
+
+type EvalSession = Awaited<ReturnType<typeof createAgentSession>>["session"];
+
+interface EvalRunResult {
+	output: string;
+	metrics: {
+		provider: string;
+		model: string;
+		inputTokens: number;
+		outputTokens: number;
+		totalTokens: number;
+	};
+}
+
+async function cleanupEvalResources(session: EvalSession | undefined, removeRoot: () => Promise<void>): Promise<void> {
+	const failures: unknown[] = [];
+	if (session) {
+		try {
+			session.dispose();
+		} catch (error) {
+			failures.push(error);
+		}
+	}
+	try {
+		await removeRoot();
+	} catch (error) {
+		failures.push(error);
+	}
+	if (failures.length === 1) {
+		throw failures[0];
+	}
+	if (failures.length > 1) {
+		throw new AggregateError(failures, "Pi eval session disposal and temporary directory cleanup failed.");
+	}
+}
+
+function getRequiredModelSelection(): { provider: string; model: string } {
+	const provider = process.env.PI_PROVIDER?.trim();
+	const model = process.env.PI_MODEL?.trim();
+	if (!provider || !model) {
+		throw new Error("PI_PROVIDER and PI_MODEL must both be set for eval runs.");
+	}
+	return { provider, model };
+}
+
+async function createPiEvalAgent({ context }: PiAiCreateAgentArgs<string>) {
+	const signal = context.signal;
+	signal?.throwIfAborted();
+	const selection = getRequiredModelSelection();
+	const modelRuntime = await ModelRuntime.create();
+	signal?.throwIfAborted();
+	const model = modelRuntime.getModel(selection.provider, selection.model);
+	if (!model) {
+		throw new Error(`Eval model not found: ${selection.provider}/${selection.model}`);
+	}
+	const auth = await modelRuntime.checkAuth(selection.provider);
+	signal?.throwIfAborted();
+	if (!auth) {
+		throw new Error(`No authentication configured for eval model ${selection.provider}/${selection.model}.`);
+	}
+
+	const root = await mkdtemp(join(tmpdir(), "pi-eval-"));
+	const cwd = join(root, "workspace");
+	const agentDir = join(root, "agent");
+	const cleanup = () => rm(root, { recursive: true, force: true });
+	let session: EvalSession | undefined;
+	try {
+		signal?.throwIfAborted();
+		await Promise.all([mkdir(cwd), mkdir(agentDir)]);
+		signal?.throwIfAborted();
+
+		const settingsManager = SettingsManager.inMemory();
+		const resourceLoader = new DefaultResourceLoader({
+			cwd,
+			agentDir,
+			settingsManager,
+			noExtensions: true,
+			noSkills: true,
+			noPromptTemplates: true,
+			noThemes: true,
+			noContextFiles: true,
+		});
+		await resourceLoader.reload();
+		signal?.throwIfAborted();
+
+		session = (
+			await createAgentSession({
+				cwd,
+				agentDir,
+				model,
+				modelRuntime,
+				thinkingLevel: "off",
+				noTools: "all",
+				resourceLoader,
+				settingsManager,
+				sessionManager: SessionManager.inMemory(cwd),
+			})
+		).session;
+		signal?.throwIfAborted();
+	} catch (error) {
+		try {
+			await cleanupEvalResources(session, cleanup);
+		} catch (cleanupError) {
+			throw new AggregateError([error, cleanupError], "Pi eval setup failed and cleanup also failed.");
+		}
+		throw error;
+	}
+	if (!session) {
+		throw new Error("Pi eval session setup completed without a session.");
+	}
+
+	return {
+		agent: session.agent,
+		async run(input: string, runtime: PiAiRuntime<PiAiToolset<string>, string>) {
+			const abort = () => void session.abort();
+			runtime.signal?.addEventListener("abort", abort, { once: true });
+			let result: EvalRunResult;
+			try {
+				runtime.signal?.throwIfAborted();
+				await session.prompt(input);
+				const assistant = session.messages
+					.slice()
+					.reverse()
+					.find((message) => message.role === "assistant");
+				if (!assistant) {
+					throw new Error("Pi eval did not produce an assistant message.");
+				}
+				if (assistant.stopReason !== "stop") {
+					throw new Error(assistant.errorMessage ?? `Pi eval stopped unexpectedly: ${assistant.stopReason}`);
+				}
+
+				const output = session.getLastAssistantText();
+				if (!output) {
+					throw new Error("Pi eval produced an empty assistant response.");
+				}
+
+				const stats = session.getSessionStats();
+				runtime.events.assistant(output, {
+					provider: model.provider,
+					model: model.id,
+				});
+				result = {
+					output,
+					metrics: {
+						provider: model.provider,
+						model: model.id,
+						inputTokens: stats.tokens.input,
+						outputTokens: stats.tokens.output,
+						totalTokens: stats.tokens.total,
+					},
+				};
+			} catch (error) {
+				runtime.signal?.removeEventListener("abort", abort);
+				try {
+					await cleanupEvalResources(session, cleanup);
+				} catch (cleanupError) {
+					throw new AggregateError([error, cleanupError], "Pi eval run failed and cleanup also failed.");
+				}
+				throw error;
+			}
+			runtime.signal?.removeEventListener("abort", abort);
+			await cleanupEvalResources(session, cleanup);
+			return result;
+		},
+	};
+}
+
+export const piCodingAgentHarness = piAiHarness({
+	name: "pi-coding-agent",
+	agent: createPiEvalAgent,
+});
diff --git a/packages/evals/tsconfig.json b/packages/evals/tsconfig.json
new file mode 100644
index 000000000..54e7cdf3f
--- /dev/null
+++ b/packages/evals/tsconfig.json
@@ -0,0 +1,10 @@
+{
+	"extends": "../../tsconfig.json",
+	"compilerOptions": {
+		"noEmit": true,
+		"module": "NodeNext",
+		"moduleResolution": "NodeNext",
+		"types": ["node", "vitest"]
+	},
+	"include": ["src/**/*.ts"]
+}
diff --git a/packages/evals/vitest.config.ts b/packages/evals/vitest.config.ts
new file mode 100644
index 000000000..6e7c6aa1d
--- /dev/null
+++ b/packages/evals/vitest.config.ts
@@ -0,0 +1,31 @@
+import { fileURLToPath } from "node:url";
+import { defineConfig, mergeConfig } from "vitest/config";
+import baseConfig, { workspaceSourcePaths } from "../../vitest.base.ts";
+
+const piAiCompatIndex = fileURLToPath(new URL("./src/pi-ai-compat.ts", import.meta.url));
+
+export default mergeConfig(
+	baseConfig,
+	defineConfig({
+		test: {
+			environment: "node",
+			fileParallelism: false,
+			include: ["src/**/*.eval.ts"],
+			testTimeout: 120000,
+			hookTimeout: 30000,
+			reporters: ["vitest-evals/reporter"],
+			server: {
+				deps: {
+					inline: [/@vitest-evals\/harness-pi-ai/],
+				},
+			},
+		},
+		resolve: {
+			alias: [
+				{ find: /^@earendil-works\/pi-coding-agent$/, replacement: workspaceSourcePaths.codingAgentIndex },
+				{ find: /^@mariozechner\/pi-ai$/, replacement: piAiCompatIndex },
+				{ find: /^@mariozechner\/pi-agent-core$/, replacement: workspaceSourcePaths.agentIndex },
+			],
+		},
+	}),
+);
diff --git a/scripts/sync-versions.js b/scripts/sync-versions.js
index 88ebdf876..01291d810 100644
--- a/scripts/sync-versions.js
+++ b/scripts/sync-versions.js
@@ -1,29 +1,47 @@
 #!/usr/bin/env node
 
 /**
- * Syncs all non-private workspace package dependency versions to match their current versions.
- * This ensures release packages, including unpublished packages, use lockstep versioning.
+ * Validates lockstep versions for published packages, then synchronizes
+ * internal dependency versions in all workspace packages, including private ones.
  */
 
 import { readFileSync, writeFileSync } from "node:fs";
 import { join } from "node:path";
 import { findPackageDirectories } from "./package-workspaces.mjs";
 
-const packages = findPackageDirectories()
+const GENERATED_PACKAGE_SUFFIXES = [join("coding-agent", "install-lock")];
+
+function synchronizedDependencyVersion(dependencyName, currentSpecifier, versionMap) {
+	const directVersion = versionMap.get(dependencyName);
+	if (directVersion) {
+		return `^${directVersion}`;
+	}
+	if (typeof currentSpecifier !== "string" || !currentSpecifier.startsWith("npm:")) {
+		return null;
+	}
+
+	const separator = currentSpecifier.lastIndexOf("@");
+	const aliasPackage = currentSpecifier.slice("npm:".length, separator);
+	const aliasVersion = versionMap.get(aliasPackage);
+	return aliasVersion ? `npm:${aliasPackage}@${aliasVersion}` : null;
+}
+
+const packageRoot = process.argv[2] ?? "packages";
+const workspacePackages = findPackageDirectories(packageRoot)
+	.filter((directory) => !GENERATED_PACKAGE_SUFFIXES.some((suffix) => directory.endsWith(suffix)))
 	.map((directory) => {
 		const path = join(directory, "package.json");
 		return { data: JSON.parse(readFileSync(path, "utf8")), path };
-	})
-	.filter((pkg) => pkg.data.private !== true);
-
-const versionMap = new Map(packages.map((pkg) => [pkg.data.name, pkg.data.version]));
+	});
+const publishedPackages = workspacePackages.filter((pkg) => pkg.data.private !== true);
+const versionMap = new Map(workspacePackages.map((pkg) => [pkg.data.name, pkg.data.version]));
 
 console.log("Current versions:");
-for (const [name, version] of [...versionMap].sort(([a], [b]) => a.localeCompare(b))) {
-	console.log(`  ${name}: ${version}`);
+for (const pkg of [...publishedPackages].sort((a, b) => a.data.name.localeCompare(b.data.name))) {
+	console.log(`  ${pkg.data.name}: ${pkg.data.version}`);
 }
 
-const versions = new Set(versionMap.values());
+const versions = new Set(publishedPackages.map((pkg) => pkg.data.version));
 if (versions.size > 1) {
 	console.error("\nERROR: Not all non-private packages have the same version.");
 	console.error("Expected lockstep versioning. Run one of:");
@@ -36,37 +54,33 @@ if (versions.size > 1) {
 console.log("\nAll non-private packages are at the same version (lockstep).");
 
 let totalUpdates = 0;
-for (const pkg of packages) {
-	let updated = false;
-
+const updatedPackages = new Set();
+for (const pkg of workspacePackages) {
 	for (const dependencyType of ["dependencies", "devDependencies"]) {
 		const dependencies = pkg.data[dependencyType];
 		if (!dependencies) {
 			continue;
 		}
 
-		for (const [dependencyName, currentVersion] of Object.entries(dependencies)) {
-			const dependencyVersion = versionMap.get(dependencyName);
-			if (!dependencyVersion) {
-				continue;
-			}
-
-			const newVersion = `^${dependencyVersion}`;
-			if (currentVersion === newVersion) {
+		for (const [dependencyName, currentSpecifier] of Object.entries(dependencies)) {
+			const newSpecifier = synchronizedDependencyVersion(dependencyName, currentSpecifier, versionMap);
+			if (!newSpecifier || currentSpecifier === newSpecifier) {
 				continue;
 			}
 
 			console.log(`\n${pkg.data.name}:`);
-			console.log(`  ${dependencyName}: ${currentVersion} → ${newVersion}${dependencyType === "devDependencies" ? " (devDependencies)" : ""}`);
-			dependencies[dependencyName] = newVersion;
-			updated = true;
+			console.log(
+				`  ${dependencyName}: ${currentSpecifier} → ${newSpecifier}${dependencyType === "devDependencies" ? " (devDependencies)" : ""}`,
+			);
+			dependencies[dependencyName] = newSpecifier;
+			updatedPackages.add(pkg);
 			totalUpdates++;
 		}
 	}
+}
 
-	if (updated) {
-		writeFileSync(pkg.path, `${JSON.stringify(pkg.data, null, "\t")}\n`);
-	}
+for (const pkg of updatedPackages) {
+	writeFileSync(pkg.path, `${JSON.stringify(pkg.data, null, "\t")}\n`);
 }
 
 if (totalUpdates === 0) {
diff --git a/scripts/sync-versions.test.mjs b/scripts/sync-versions.test.mjs
new file mode 100644
index 000000000..433ccb51c
--- /dev/null
+++ b/scripts/sync-versions.test.mjs
@@ -0,0 +1,75 @@
+import assert from "node:assert/strict";
+import { spawnSync } from "node:child_process";
+import { mkdtemp, mkdir, readFile, rm, writeFile } from "node:fs/promises";
+import { tmpdir } from "node:os";
+import { join } from "node:path";
+import { fileURLToPath } from "node:url";
+import test from "node:test";
+
+const syncVersionsScript = fileURLToPath(new URL("./sync-versions.js", import.meta.url));
+
+async function writeManifest(root, relativeDirectory, manifest) {
+	const directory = join(root, relativeDirectory);
+	await mkdir(directory, { recursive: true });
+	await writeFile(join(directory, "package.json"), `${JSON.stringify(manifest, null, "\t")}\n`);
+}
+
+async function readManifest(root, relativeDirectory) {
+	return JSON.parse(await readFile(join(root, relativeDirectory, "package.json"), "utf8"));
+}
+
+function runSyncVersions(root) {
+	return spawnSync(process.execPath, [syncVersionsScript, join(root, "packages")], {
+		cwd: root,
+		encoding: "utf8",
+	});
+}
+
+test("synchronizes private dependencies without touching generated manifests or relaxing published lockstep", async () => {
+	const root = await mkdtemp(join(tmpdir(), "pi-sync-versions-"));
+	try {
+		await writeManifest(root, "packages/ai", {
+			name: "@earendil-works/pi-ai",
+			version: "2.0.0",
+		});
+		await writeManifest(root, "packages/coding-agent", {
+			name: "@earendil-works/pi-coding-agent",
+			version: "2.0.0",
+		});
+		await writeManifest(root, "packages/evals", {
+			name: "@earendil-works/pi-evals",
+			version: "9.9.9",
+			private: true,
+			dependencies: {
+				"@earendil-works/pi-coding-agent": "^1.0.0",
+				"@mariozechner/pi-ai": "npm:@earendil-works/pi-ai@1.0.0",
+			},
+		});
+		await writeManifest(root, "packages/coding-agent/install-lock", {
+			name: "generated-install-lock",
+			version: "0.0.0",
+			private: true,
+			dependencies: {
+				"@earendil-works/pi-coding-agent": "^1.0.0",
+			},
+		});
+
+		const result = runSyncVersions(root);
+		assert.equal(result.status, 0, result.stderr);
+
+		const evalsManifest = await readManifest(root, "packages/evals");
+		assert.equal(evalsManifest.dependencies["@earendil-works/pi-coding-agent"], "^2.0.0");
+		assert.equal(evalsManifest.dependencies["@mariozechner/pi-ai"], "npm:@earendil-works/pi-ai@2.0.0");
+		const generatedManifest = await readManifest(root, "packages/coding-agent/install-lock");
+		assert.equal(generatedManifest.dependencies["@earendil-works/pi-coding-agent"], "^1.0.0");
+
+		await writeManifest(root, "packages/ai", {
+			name: "@earendil-works/pi-ai",
+			version: "3.0.0",
+		});
+		const lockstepFailure = runSyncVersions(root);
+		assert.equal(lockstepFailure.status, 1, lockstepFailure.stderr);
+	} finally {
+		await rm(root, { recursive: true, force: true });
+	}
+});
diff --git a/vitest.base.ts b/vitest.base.ts
new file mode 100644
index 000000000..1ede2a2f3
--- /dev/null
+++ b/vitest.base.ts
@@ -0,0 +1,28 @@
+import { fileURLToPath } from "node:url";
+import { defineConfig } from "vitest/config";
+
+export const workspaceSourcePaths = {
+	aiIndex: fileURLToPath(new URL("./packages/ai/src/index.ts", import.meta.url)),
+	aiCompat: fileURLToPath(new URL("./packages/ai/src/compat.ts", import.meta.url)),
+	aiOAuth: fileURLToPath(new URL("./packages/ai/src/oauth.ts", import.meta.url)),
+	aiProviders: fileURLToPath(new URL("./packages/ai/src/providers", import.meta.url)),
+	agentIndex: fileURLToPath(new URL("./packages/agent/src/index.ts", import.meta.url)),
+	codingAgentIndex: fileURLToPath(new URL("./packages/coding-agent/src/index.ts", import.meta.url)),
+	tuiIndex: fileURLToPath(new URL("./packages/tui/src/index.ts", import.meta.url)),
+} as const;
+
+export default defineConfig({
+	resolve: {
+		alias: [
+			{ find: /^@earendil-works\/pi-ai$/, replacement: workspaceSourcePaths.aiIndex },
+			{ find: /^@earendil-works\/pi-ai\/compat$/, replacement: workspaceSourcePaths.aiCompat },
+			{ find: /^@earendil-works\/pi-ai\/oauth$/, replacement: workspaceSourcePaths.aiOAuth },
+			{
+				find: /^@earendil-works\/pi-ai\/providers\/(.+)$/,
+				replacement: `${workspaceSourcePaths.aiProviders}/$1.ts`,
+			},
+			{ find: /^@earendil-works\/pi-agent-core$/, replacement: workspaceSourcePaths.agentIndex },
+			{ find: /^@earendil-works\/pi-tui$/, replacement: workspaceSourcePaths.tuiIndex },
+		],
+	},
+});