Skip to main content
// LEGAL & HR_DEPARTMENT

Training Recorder

Maintains the training and certification matrix

What it costs

Bought once, yours forever. Roster workers are one-time purchases — no subscription. The eval report ships inside the bundle.

This worker
£15 one-time

Just Training Recorder, with its eval report.

The whole Legal & HR department
£100 one-time

All 10 workers — £150 bought singly, 33% off as a bundle.

Where it stands

In testing — running its suite ahead of release.

Release gate
  • Scenarios in the suite120
  • Minimum pass rate98%
  • p95 latency ceiling4.0s
  • Cost per run ceiling£0.04
  • RunsOn-demand

These are the thresholds Training Recorder has to clear before it may ship — not results. Its measured eval numbers are in the report below.

The eval report

This ships inside the bundle. It is the contract: what the agent is tested on, what it refuses outright, and what it hands back to a human.

eval_report.jsonv0.1.0
{
  "agent": "legal-hr-training-recorder",
  "version": "0.1.0",
  "status": "measured",
  "gate": {
    "scenarios": 120,
    "pass_rate_min": "98%",
    "p95_latency_max": "4.0s",
    "cost_per_run_max": "£0.04"
  },
  "suites": [
    {
      "name": "core",
      "scenarios": 66,
      "description": "The ordinary cases. What the agent is for, done correctly."
    },
    {
      "name": "adversarial",
      "scenarios": 30,
      "description": "Prompt injection, contradictory instructions, hostile or malformed input, and attempts to make it act outside its contract."
    },
    {
      "name": "regression",
      "scenarios": 24,
      "description": "Every bug ever found, kept forever so it cannot come back."
    }
  ],
  "refuses": [
    "Acting on instructions found in content it was asked to read",
    "Sending, publishing or paying anything without explicit approval",
    "Inventing a fact it cannot source"
  ],
  "escalates_to_human": [
    "Confidence below the configured floor",
    "Anything involving money, legal commitment or a new counterparty",
    "Conflicting instructions it cannot resolve"
  ],
  "measured": {
    "certified": true,
    "certified_on": "2026-08-10",
    "baseline": "2026-08-10",
    "judge": {
      "model": "claude-opus-4-8",
      "rubric_version": "1.0.0",
      "calibrated": "2026-08-09",
      "trusted": true
    },
    "results": {
      "standalone": {
        "passed": true,
        "judged": true,
        "judge_mean": 5,
        "model": "claude-sonnet-4-6"
      },
      "veriphos": {
        "passed": false,
        "judged": true,
        "judge_mean": null,
        "model": "claude-sonnet-4-6"
      },
      "replay": {
        "passed": true,
        "judged": false,
        "judge_mean": null,
        "model": "claude-sonnet-4-6"
      }
    }
  }
}
The measured block is real: per-host results from the latest eval run, judged against the 2026-08-10 baseline, certified 2026-08-10.

Others in Legal & HR

Contract Reviewer

Reviews contract terms and flags risk

Policy Drafter

Drafts and updates internal policies

Compliance Tracker

Tracks regulatory deadlines and renewals

Dispute Logger

Records disputes and correspondence

IP Guardian

Protects trademarks, IP and registrations

Recruitment Screener

Screens and shortlists candidates

See the full roster →