Critical Agent

1. What Critical Agent Does

Critical Agent is seeded into a quantum analysis to refuse to settle a claim unless the pooled evidence corroborates it from multiple independent sources. Where Convergent argues the majority reading and Divergent argues the minority reading, Critical's job is to catch claims that only look well-evidenced, a single loud data point dressed up as consensus.

It pulls the same fused evidence pool as Convergent and Divergent but applies a corroboration requirement before it will commit to a settled verdict: at least two independent sources must point the same direction, or it must return insufficient, however confident a single source looks.

2. Where It Sits in the Tao Status Economy

Critical Agent is a consumer of miner-contributed LLM keys, preferring a pooled miner key on each attempt and falling back to its own static key, invisible to scoring, only when contribution is disabled, the pool declines, or the vendor is unsupported.

3. The Turn Pipeline

• No evidence at all across every pooled source: the agent declines rather than guessing.
• Up to five attempts are made, each with a freshly acquired key.

4. Quality Grading

5. Reporting

One report is filed per turn, feeding directly into the reporting miner's scoring window:

{
  "hotkey": "5F...",
  "success": true,
  "latency_ms": 1734.6,
  "error_category": null,
  "quality_score": 0.6,
  "lease_token": "the token from this attempt's acquire grant"
}

• success is false only on a raised exception during the LLM call.
• A report without a valid, unconsumed lease_token is rejected and never recorded.

6. Simulated Example

A quantum analysis turn on the claim: "Subnet 40 is gaining developer traction."

1. 1

   Pooled retrieval

   GitHub shows a genuine commit spike, 410 commits against a 90-commit baseline, but Discord and Twitter both show flat activity, and wallet data has no relevant signal for developer traction. Only one of four applicable sources corroborates the claim.

2. 2

   Attempt 1

   The first miner's key returns a SUPPORTED verdict citing only the commit jump. Single-source confidence on a multi-source claim is an overclaim, reported at quality 0.2.

3. 3

   Attempt 2

   A second miner's key returns an insufficient verdict that cites the commit jump but also notes the flat Discord and Twitter activity. This correctly withholds a settled verdict given only one-of-four corroboration, and it does cite evidence rather than hedging blindly, graded 0.6.

7. Operating Limits