ワンクリックで
runtime-validation-repair-worker
Repair regressions discovered by validation during the `src/fleet_rlm` simplification mission.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Repair regressions discovered by validation during the `src/fleet_rlm` simplification mission.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Delegate recursive work to child RLM sandboxes with budget management. Use when decomposing tasks into sub-queries, fanning out batched work, managing LLM call budgets, or building parent-child RLM hierarchies.
Diagnose fleet-rlm runtime failures, API contract drift, sandbox errors, and observability issues. Use when something is broken — Daytona connection failures, websocket mismatches, escalation not triggering, budget exhaustion, or missing traces.
Optimize fleet-rlm DSPy programs and RLM skill bundles using GEPA with MLflow tracking. Use when running optimization loops, designing feedback metrics, building training datasets, or comparing runs.
Hub skill for fleet-rlm: when to use dspy.RLM vs ReAct/CodeAct and which workflow skill to load next.
Design DSPy signatures and compose runtime modules for fleet-rlm tasks. Use when creating input/output field definitions, choosing between built-in signatures, selecting execution modes, or wiring custom modules.
Fetch and inspect JavaScript-heavy pages with Playwright in a Daytona browser-capable snapshot.
| name | runtime-validation-repair-worker |
| description | Repair regressions discovered by validation during the `src/fleet_rlm` simplification mission. |
NOTE: Startup and cleanup are handled by worker-base. This skill defines the work procedure for repair/fix features created after failed validation or worker discovery.
Use this skill for features that:
agent-browser — invoke when the regression affects the browser shell, route serving, or shell navigation behavior.mission-worker-base already ran .factory/init.sh and the baseline commands.test command from .factory/services.yaml for this session. If either step did not run or failed, run it before repair work and record the result; if it still fails, return to the orchestrator without editing code.mission.md, mission AGENTS.md, and the .factory/library/* files referenced there.src/fleet_rlm/api/bootstrap_observability.py unless the repair explicitly requires touching it.make formatmake lintmake typecheck.factory/services.yaml (api_contract, websocket_contract, daytona_contract, cli_contract, or quality_gate).agent-browser against 127.0.0.1:8100 to verify the repaired flow manually.{
"salientSummary": "Repaired a websocket regression introduced by runtime simplification so basic chat streaming again emits canonical metadata and terminates cleanly. The failing websocket test now passes and the API-served shell still works.",
"whatWasImplemented": "Added a focused regression test around the broken stream metadata path, repaired the event-shaping helper to restore canonical websocket payloads, and re-ran the focused websocket contract lane plus baseline validators. The fix stayed within the existing ownership boundary and did not add new wrapper layers.",
"whatWasLeftUndone": "",
"verification": {
"commandsRun": [
{
"command": "uv run pytest -q tests/ui/ws/test_chat_stream.py::test_websocket_basic_message_flow -m 'not live_llm and not live_daytona and not benchmark'",
"exitCode": 0,
"observation": "The exact failing websocket assertion now passes."
},
{
"command": "uv run pytest -q tests/ui/ws/test_chat_stream.py tests/ui/ws/test_commands.py tests/unit/test_ws_chat_helpers.py -m 'not live_llm and not live_daytona and not benchmark'",
"exitCode": 0,
"observation": "Focused websocket contract coverage stayed green after the repair."
},
{
"command": "make format && make lint && make typecheck",
"exitCode": 0,
"observation": "Baseline validators passed after the fix."
}
],
"interactiveChecks": [
{
"action": "Loaded the API-served app shell on http://127.0.0.1:8100 with agent-browser after the websocket repair.",
"observed": "The shell still loaded and the repair did not break the browser surface."
}
]
},
"tests": {
"added": [
{
"file": "tests/ui/ws/test_chat_stream.py",
"cases": [
{
"name": "test_websocket_basic_message_flow",
"verifies": "Canonical websocket metadata and final-event behavior remain intact after simplification."
}
]
}
]
},
"discoveredIssues": []
}