Verified external Agents
0 / 1000
Agent usage evidence
Verified adoption requires a manually admitted independent operator and a matching successful external Agent execution. Admissions alone, copied IDs, failures, anonymous calls, probes, crawlers, and internal tests do not count.
Verified independent operators
0
Verified successful executions
0
Active evidence admissions
0
Admissions without successful execution do not count
Unverified installation IDs
9
Diagnostic only · 80 anonymous successes
7-day verified retention
not enough data
0/0 eligible
30-day verified retention
not enough data
0/0 eligible
30-day external success rate
80.5%
41 observations
Risk preflight — by attribution
| Scope | Evaluations | Identified installations | Allow | Review | Block | Outcome reports | Reported behavior changes |
|---|---|---|---|---|---|---|---|
| Identified external installations | 1 | 1 | 1 | 0 | 0 | 0 (0.0%) | 0 (not enough data) |
| Anonymous external | 3 | 0 | 3 | 0 | 0 | 0 (0.0%) | 0 (not enough data) |
| Internal / non-external | 4 | 0 | 3 | 0 | 1 | 2 (50.0%) | 1 (50.0%) |
| Unattributed | 0 | 0 | 0 | 0 | 0 | 0 (not enough data) | 0 (not enough data) |
| All traffic (not adoption) | 8 | 1 | 7 | 0 | 1 | 2 (25.0%) | 1 (50.0%) |
Outcomes are self-reported unless a receipt explicitly marks evidence_level=observed. Behavior changes do not establish causation, independent users, or prevented losses. Self-reported outcomes do not directly increase Trust scores.
This attribution table is diagnostic. The verified adoption cards above require separate independent-operator evidence plus successful execution.
Prediction-market preflight — by attribution
| Scope | Evaluations | Identified installations | Allow | Review | Block | Outcome reports | Reported behavior changes |
|---|---|---|---|---|---|---|---|
| Identified external installations | 2 | 1 | 0 | 2 | 0 | 0 (0.0%) | 0 (not enough data) |
| Anonymous external | 9 | 0 | 0 | 8 | 1 | 0 (0.0%) | 0 (not enough data) |
| Internal / non-external | 6 | 0 | 0 | 5 | 1 | 2 (33.3%) | 1 (50.0%) |
| Unattributed | 0 | 0 | 0 | 0 | 0 | 0 (not enough data) | 0 (not enough data) |
| All traffic (not adoption) | 17 | 1 | 0 | 15 | 2 | 2 (11.8%) | 1 (50.0%) |
Behavior outcomes are self-reported and do not prove causation, independent users, prevented losses, profitability, or prediction accuracy. Market resolution calibration is not included. Self-reported outcomes do not directly increase Trust scores.
This attribution table is diagnostic. The verified adoption cards above require separate independent-operator evidence plus successful execution.
Activation by source
| Source | Views | Installs | Initialized Agents | Prompt-opened Agents | Calling Agents | Successful Agents | Failed Agents | Call rate | Call success | Prompt→success | Activation |
|---|---|---|---|---|---|---|---|---|---|---|---|
npx-proxy |
0 | 0 | 166 | 0 | 7 | 7 | 7 | 4.2% | 100.0% | not enough data | 4.2% |
cursor |
1 | 23 | 2 | 0 | 2 | 2 | 1 | 100.0% | 100.0% | not enough data | 100.0% |
direct |
103 | 0 | 1 | 0 | 0 | 0 | 0 | 0.0% | not enough data | not enough data | 0.0% |
claude |
0 | 0 | 0 | 0 | 0 | 0 | 0 | not enough data | not enough data | not enough data | not enough data |
release-smoke |
1 | 0 | 0 | 0 | 0 | 0 | 0 | not enough data | not enough data | not enough data | not enough data |
full-audit |
0 | 0 | 0 | 0 | 0 | 0 | 0 | not enough data | not enough data | not enough data | not enough data |
official-registry |
0 | 0 | 37 | 0 | 0 | 0 | 0 | 0.0% | not enough data | not enough data | 0.0% |
mcp-inspector |
0 | 0 | 0 | 0 | 0 | 0 | 0 | not enough data | not enough data | not enough data | not enough data |
claude.acceptance |
0 | 0 | 0 | 0 | 0 | 0 | 0 | not enough data | not enough data | not enough data | not enough data |
first-shot-smoke |
7 | 0 | 0 | 0 | 0 | 0 | 0 | not enough data | not enough data | not enough data | not enough data |
codex.acceptance |
0 | 0 | 0 | 0 | 0 | 0 | 0 | not enough data | not enough data | not enough data | not enough data |
cursor.acceptance |
0 | 0 | 0 | 0 | 0 | 0 | 0 | not enough data | not enough data | not enough data | not enough data |
Tool reliability — last 30 days
| Tool | Calls | Agents | Success | P95 ms | Last observed |
|---|---|---|---|---|---|
understand_webpage |
3 | 3 | 0.0% | 1 | 2026-10-04T02:17:11.938Z |
evaluate_prediction_market |
3 | 3 | 0.0% | 1 | 2026-10-04T02:17:13.392Z |
verify_web |
3 | 3 | 100.0% | 9 | 2026-10-04T02:17:12.810Z |
search_tools |
21 | 0 | 90.5% | 1825 | 2026-10-08T05:46:08.729Z |
get_capability_graph |
7 | 0 | 100.0% | 1344 | 2026-10-08T02:01:02.512Z |
list_capabilities |
4 | 0 | 100.0% | 1203 | 2026-10-06T11:34:14.292Z |
Canonical errors — last 30 days
| Error | Events |
|---|---|
tool_execution_failed | 3 |
unsafe_url | 3 |
unknown | 2 |
Generated 2026-10-08T09:21:33.932Z. Raw Agent IDs, operator IDs, evidence references, session IDs, prompts, arguments and results are never shown.