mm_verify_compute_confidence
Pure per-run confidence score derived from convergence state + tier counts + adversarial-bank signal. Returns { bucket, numeric, rationale[] } where bucket ∈ {high, medium, low, unstable} is authoritative and numeric ∈ [0,1] is a visualization hint. Mirrors Rust score in commands/verify/confidence.rs byte-for-byte. Used by /mm:verify-spec --iterative IStep 4 to emit a one-line confidence summary.
At a Glance
Section titled “At a Glance”- Category: Plugin Management
- Source: Built-in
- Required parameters: 9
- Total parameters: 9
Parameters
Section titled “Parameters”| Parameter | Type | Required | Description |
|---|---|---|---|
runs_completed | number | Yes | Total runs completed so far (>=0). |
consecutive_stable | number | Yes | Running stability counter from mm_verify_convergence_step. |
decision | string (enum: continue, converged, did_not_converge) | Yes | Final decision from mm_verify_convergence_step for this run. |
t1_count | number | Yes | Count of T1 findings on the final run. |
t2_count | number | Yes | Count of T2 findings on the final run. |
t3_count | number | Yes | Count of T3 findings on the final run. |
t4_count | number | Yes | Count of T4 (advisory) findings on the final run. |
adversarial_null_count | number | Yes | Number of the 3 adversarial prompts that returned null_result this run. |
adversarial_malformed | number | Yes | Number of adversarial prompts that produced malformed output. Any non-zero value forces bucket=unstable. |
Example
Section titled “Example”{ "tool": "mm_verify_compute_confidence", "arguments": { "runs_completed": 1, "consecutive_stable": 1, "decision": "continue", "t1_count": 10, "t2_count": 10, "t3_count": 10, "t4_count": 10, "adversarial_null_count": 10, "adversarial_malformed": 1 }}Related Tools
Section titled “Related Tools”mm_list_plugins- List all loaded plugins with their id, name, version, category, tier, and too…mm_get_plugin_info- Get detailed information about a specific plugin including its tools, version…mm_push_team_checkin- Push a team check-in to the cloud. Scans local specs and features, captures c…mm_fetch_team_checkins- Fetch all team members’ check-ins from the cloud. Returns display names, bran…mm_claim_spec- Claim a spec from a teammate. Records the claim in the cloud (first-claim-win…