Files
openclaw/docs/concepts/personal-agent-benchmark-pack.md
2026-05-17 02:56:53 +08:00

2.7 KiB

summary, read_when, title
summary read_when title
Local qa-channel scenarios for privacy-preserving personal assistant workflow checks.
Running local personal agent reliability checks
Extending the repo-backed QA scenario catalog
Verifying reminder, reply, memory, redaction, and safe tool followthrough behavior
Personal agent benchmark pack

The Personal Agent Benchmark Pack is a small repo-backed QA scenario pack for local personal assistant workflows. It is not a generic model benchmark and it does not require a new runner. The pack reuses the private QA stack described in QA overview, the synthetic QA channel, and the existing qa/scenarios markdown catalog.

The first pack is intentionally narrow:

  • fake personal reminders through local cron delivery
  • fake DM and thread reply routing through qa-channel
  • fake preference recall from the temporary QA workspace memory files
  • fake secret no-echo checks
  • safe read-backed tool followthrough after a short approval-style turn

Scenarios

The machine-readable pack metadata lives in extensions/qa-lab/src/scenario-packs.ts. The initial pack does not add a CLI pack selector, so run the scenarios explicitly:

OPENCLAW_ENABLE_PRIVATE_QA_CLI=1 pnpm openclaw qa suite \
  --provider-mode mock-openai \
  --scenario personal-reminder-roundtrip \
  --scenario personal-channel-thread-reply \
  --scenario personal-memory-preference-recall \
  --scenario personal-redaction-no-secret-leak \
  --scenario personal-tool-safety-followthrough \
  --concurrency 1

The pack is designed for qa-channel with mock-openai or another local QA provider lane. It should not be pointed at live chat services or real personal accounts.

Privacy Model

The scenarios use only fake users, fake preferences, fake secrets, and the temporary QA gateway workspace created by the suite. They must not read or write real OpenClaw user memory, sessions, credentials, launch agents, global configs, or live gateway state.

Artifacts stay under the existing QA suite artifact directory and should be treated like test output. Redaction checks use fake markers so failures are safe to inspect and file in issues.

Extending The Pack

Add new cases under qa/scenarios/personal/, then add the scenario id to QA_PERSONAL_AGENT_SCENARIO_IDS. Keep each case small, local, deterministic in mock-openai, and focused on one personal assistant behavior.

Good follow-up candidates:

  • approval denial correctness
  • multi-step task ledger assertions
  • redacted trajectory export checks
  • local-only plugin workflow checks

Avoid adding a new runner, plugin, dependency, live transport, or model judge until the scenario catalog has enough stable cases to justify that surface.