operant-mcp
Read-only MCP server for the OPERANT AI operating-agent calibration benchmark.
Connect it
Settings → Connectors → Add custom connector
URL: https://operant-mcp.saagar210.workers.dev/mcpSettings → Apps & Connectors → Advanced → Developer mode
Create connector with URL: https://operant-mcp.saagar210.workers.dev/mcp{
"mcpServers": {
"operant-mcp": {
"url": "https://operant-mcp.saagar210.workers.dev/mcp"
}
}
}claude mcp add --transport http operant-mcp https://operant-mcp.saagar210.workers.dev/mcp{
"servers": {
"operant-mcp": {
"type": "http",
"url": "https://operant-mcp.saagar210.workers.dev/mcp"
}
}
}{
"mcpServers": {
"operant-mcp": {
"command": "npx",
"args": [
"-y",
"saagar-operant-mcp@0.1.0"
]
}
}
}claude mcp add operant-mcp -- npx -y saagar-operant-mcp@0.1.05 tools, verified live
get_resultsread-onlyReturn retained model calculation profiles plus generated_at, claim_status, claims_at_risk, and the bounded evidence_binding. These rows are not durable named-model performance claims. Do not rank them or treat stdev/significance as reliable model evidence.
compare_modelsread-onlyPlace two retained calculation profiles side by side by display_name substring. Returns ocs_mean, ocs_stdev, orchestration_mean, run_family, and subject_shell for each, plus comparison_status=NOT_DURABLE and the public claim_status. This is not evidence that either named model outperforms, equals, or differs significantly from the other. If a name is ambiguous or not found, returns an error listin
get_methodologyread-onlyReturn the benchmark design: the 4 axes, the OCS formula (Youden's J), the 5 decision labels (PROCEED, PROCEED_SANCTIONED, REFUSE, ESCALATE, REROUTE), scoring blocks, and a concise what-it-measures / what-it-doesn't summary. Sourced from the baked case-file metadata.
list_casesread-onlyReturn case metadata (no full task prompts): pair_id/id, axis, tier, grounding, and side indicators (malign/benign for axes 1/2/4; null for axis 3). Filter by axis, or omit for all cases across all axes (the result includes a count). Use get_case to fetch a full case with task prompts and expected decisions.
get_caseread-onlyReturn the full case for a given pair_id (axes 1/2/4) or id (axis 3): malign and benign task prompts, expected decisions, grounding rationale, and bypass patterns. Axis 3 cases are single (unmatched) and use an 'id' field instead of 'pair_id'. Use list_cases to browse available ids.
Alternatives in Developer Tools
Compare →MGMCP GrafanaOfficialA
An MCP server giving access to Grafana dashboards, data and more.
MDMCP Devkit ServerOfficialA
Provides AI assistants with direct access to Mapbox developer APIs and documentation.
Front-End ChecklistA
Review frontend code and live pages against 386 quality-gated web development rules.
PMPostHog MCP ServerOfficialA
Official PostHog MCP Server for product analytics, feature flags, experiments, and more.
MAMakeOfficialA
Build, run and debug Make automation scenarios.
ClerkOfficialA
Access Clerk authentication docs, SDK snippets, and quickstart guides