Learning Resources

never86.ai

Provides public Never86 operator OS resources, rulebooks, audit logic, AEO answers, and free agent listings.

1 endpoint16 known toolsFirst detected September 26, 2026Last detected September 26, 2026

ENDPOINT 1

https://www.never86.ai/api/mcp

No auth detected

MCP server metadata

Name: never86
Version: 3.1.0

Capabilities

toolsresourcesprompts

Known tools 16

get_operator_system

Return the versioned public Never86 operator OS pack (loop, routines, agents, truth gates, safety, private-store boundary).

Inferred read-only

get_operator_logic

Fetch one public Never86 rulebook domain, or the full public logic bundle when domain is omitted or "all".

Inferred read-only

get_3p_audit_logic

Return public marketplace 3P audit rules: evidence ladder, formulas, DoorDash mappings, claim boundaries, and payout tolerance.

Inferred read-only

list_answers

List published public AEO answers (slug, title, question, audience, URL).

Inferred read-only

list_free_agents

List CSV-runnable free agents (slug, name, seat, headline, catches, sample signal, public URLs).

Inferred read-only

list_agent_jobs

List the public one-agent-one-job registry: supervisor + labor/vendor/voids/action-shift/memory, plus company roles and free-agent product catalog.

Inferred read-only

list_specialists

List domain specialists (labor, vendor, voids, action-shift, memory) with one job each, allowed tools, and MCP resource URIs.

Inferred read-only

analyze_labor

Analyze an operator-provided labor or time-clock CSV for schedule-versus-actual drift, early clock-ins, late clock-outs, overtime concentration, and possible shifts with no attached sales.

Inferred read-only

analyze_beverage

Analyze an operator-provided beverage / bar CSV for pour-vs-inventory or package cost patterns.

Inferred read-only

convert_uom

Convert verified restaurant package sizes into pour/portion units (bottle mL → fl oz, keg gal → fl oz, pours per package, cost per pour).

Inferred read-only

ask_pour_standards

Interview pack for THIS restaurant unit’s drink pour sizes.

Inferred read-only

declare_pour_standards

Validate operator-answered house pour lines for one unit (e.g.

Inferred read-only

ask_fountain_standards

Interview pack for fountain / bag-in-box gun soda at THIS unit: BIB syrup gallons, invoice $, mix ratio (e.g.

Potential side effects

analyze_recipe_cost

Cost a plate (mode=recipe), a liquor drink (mode=drink_recipe), or a fountain+spirit drink (mode=fountain_spirit_drink).

Inferred read-only

analyze_vendor_prices

Read an operator-provided vendor invoice or purchasing CSV SKU by SKU and surface price increases greater than 5% between the two latest periods.

Potential side effects

build_action_shift

Turn one restaurant's prior-day close into no more than three prioritized actions plus a night proof checklist.

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.never86]
url = "https://www.never86.ai/api/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "never86": {
      "type": "http",
      "url": "https://www.never86.ai/api/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: never86
Remote MCP URL: https://www.never86.ai/api/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "never86": {
      "url": "https://www.never86.ai/api/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "never86": {
      "type": "http",
      "url": "https://www.never86.ai/api/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "never86",
  "transport": "streamable-http",
  "url": "https://www.never86.ai/api/mcp"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Loading Trust v2 evidence…

Checking the associated registrable domain. The BuiltWith key remains server-side.

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.