MCP tool
probierz_run
EXECUTE a target end-to-end, capture evidence, analyze it, and dispatch a bounded Brama repair worker when it fails. Heavy + side-effecting: needs the target toolchain; noRepair=true records without repair.
Input schema#
{
"type": "object",
"properties": {
"target": {
"type": "string",
"description": "One of web, electron, mobile:ios, mobile:android, desktop:mac, desktop:win."
},
"record": {
"type": "boolean",
"description": "Force video + trace + screenshot capture on."
},
"appId": {
"type": "string",
"description": "Product identifier used in the run-scoped artifact path and manifest."
},
"env": {
"type": "object",
"description": "Condition env vars, e.g. { BASE_URL, APP_IOS, PROBIERZ_LOCALE, PROBIERZ_COLOR_SCHEME }."
},
"timeoutMs": {
"type": "number",
"description": "Kill the run after this many ms (default 20 min)."
},
"resourceWaitMs": {
"type": "number",
"description": "Wait this long for a busy device/port lease; 0 fails fast."
},
"frames": {
"type": "number",
"description": "Extract this many frames per recorded video (needs ffmpeg)."
},
"analyze": {
"type": "boolean",
"description": "Analyze the report after the run (default true)."
},
"force": {
"type": "boolean",
"description": "Skip the preflight gate and spawn even if the toolchain looks incomplete."
},
"spec": {
"type": "string",
"description": "Run only this one spec (path/substring), e.g. packages/mobile/test/specs/byk.e2e.ts, to scope the run to a single app's suite."
},
"noRepair": {
"type": "boolean",
"description": "Record a failed run without dispatching Brama."
}
},
"required": [
"target"
]
}Source snapshot#
Generated from `probierz-mcp` at https://github.com/wisent-ai/probierz.git revision `8ca8c1f19a5eb137164120d54f80405b0d0dbf14`.
The server reports version `0.1.0`; binary SHA-256: `72425bce4d10d4373586fb2ea7deb40be44bf277964e14aa53e52c9216e60bf4`; tools/list SHA-256: `856670843b435996d8cd5e1542c3001560c0cf11baf48a2c528f68e31600fdde`.