Imported from openclaw-easy/openclaw-easy-desktop (openclaw/src/agents/embedded-agent-runner/run/AGENTS.md). Install upstream with npx skills add openclaw-easy/openclaw-easy-desktop --skill run. Copyright stays with the author.

Embedded Runner Test Performance

Name: run
Author: openclaw-easy

The embedded attempt runner is one of the most expensive agent test surfaces. Use full-runner tests only when the behavior truly requires the runner.

Guardrails

Prefer focused helper tests for prompt assembly, runtime-context construction, cache metadata, token accounting, and maintenance decision logic.
Keep full runEmbeddedAttempt coverage for cross-component behavior that cannot be proven through helpers, not for a single derived field.
When extracting a helper from runner logic, make production call that helper directly, then test the helper. Avoid test-only copies of runner behavior.
Preserve context-engine coverage for sessionKey, sessionFile, token budget, current token count, prompt cache, and routing fields when slimming tests.
Treat a standalone full-runner test above a few seconds as suspect. First ask whether the proof can move to a production helper plus one cheap integration smoke.

Verification

For runner test slimming, run the touched helper test and the nearest two-file runner/context-engine surface.
Record Vitest duration, wall time, and RSS when the change is performance motivated.

# after Install: the listing is in your workspace's routing pool - a plan picks it for its slot curl -s -X POST https://api.opensmartroute.ai/api/v1/route -H 'Authorization: Bearer $OSR_API_KEY' -H 'Content-Type: application/json' -d '{"text": "...", "plan": true}'

{ "ocm": "1", "id": "openclaw-easy-openclaw-easy-desktop-run", "kind": "skill", "name": "run", "description": "The embedded attempt runner is one of the most expensive agent test surfaces. Use full-runner tests only when the behavior truly requires the runner.", "publisher": "openclaw-easy", "version": "1.0.0", "capabilities": { "domains": [ "coding" ], "tags": [ "skill-md", "github-agents" ], "languages": [ "en" ] }, "quality_prior": 0.6, "examples": [ "The embedded attempt runner is one of the most expensive agent test surfaces. Use full-runner tests only when the behavior truly requires the runner." ], "primary": false, "metadata": { "source": { "provider": "github-agents", "repository": "https://github.com/openclaw-easy/openclaw-easy-desktop", "path": "openclaw/src/agents/embedded-agent-runner/run/AGENTS.md", "ref": "8403220b86b69f05478eeec3114d099eb69d9a84", "url": "https://github.com/openclaw-easy/openclaw-easy-desktop/blob/8403220b86b69f05478eeec3114d099eb69d9a84/openclaw/src/agents/embedded-agent-runner/run/AGENTS.md", "key": "openclaw-easy/openclaw-easy-desktop/openclaw/src/agents/embedded-agent-runner/run/AGENTS.md" } }, "instructions": "\n\n# Embedded Runner Test Performance\n\nThe embedded attempt runner is one of the most expensive agent test surfaces.\nUse full-runner tests only when the behavior truly requires the runner.\n\n## Guardrails\n\n- Prefer focused helper tests for prompt assembly, runtime-context construction,\n cache metadata, token accounting, and maintenance decision logic.\n- Keep full `runEmbeddedAttempt` coverage for cross-component behavior that\n cannot be proven through helpers, not for a single derived fiel", "cost": { "context_tokens": 327 } }

run

About

Embedded Runner Test Performance

Guardrails

Verification

Use it

Manifest

Reviews