Datasets:
id stringlengths 16 16 | domain_id int64 1 20 | domain stringclasses 20
values | tools stringlengths 394 567 | prompt stringlengths 204 287 | response stringlengths 6.99k 8.32k | code stringlengths 4.77k 6.14k | lines_count int64 115 141 | messages listlengths 3 3 |
|---|---|---|---|---|---|---|---|---|
p16_sft_d01_0000 | 1 | Model Context Protocol (MCP) Server Protocol | [{"type": "function", "function": {"name": "mcp_server_dispatch_0", "description": "Executes standard MCP JSON-RPC 2.0 tool invocation with schema validation and capability discovery.", "parameters": {"type": "object", "properties": {"server_uri": {"type": "string"}, "method": {"type": "string", "enum": ["tools/call", ... | Implement an enterprise-grade Model Context Protocol (MCP) server class 'FastMcpEngine_000' in Python complying with JSON-RPC 2.0 standards. It must validate tool schemas, enforce session token authentication, apply resilience bounds, and output structured execution telemetry. | <think>
1. Specification Decomposition & Architectural Analysis:
- Target Domain: Model Context Protocol (MCP) Server Protocol enterprise implementation.
- Primary Execution Target: Class 'FastMcpEngine_000', Core Method: 'dispatch_tool_call'.
- Selected Data Modeling Strategy: Schema Paradigm #0.
- Standar... | import time
import json
import logging
import secrets
import hashlib
import hmac
from typing import Dict, Any, List, Optional, Tuple
from pydantic import BaseModel, Field, ConfigDict
logger = logging.getLogger("FastMcpEngine_000")
class McpProtocolException(Exception):
def __init__(self, message: str = "Model Con... | 131 | [
{
"role": "system",
"content": "You are an expert autonomous software engineer and tool-use agent capable of writing mission-critical, enterprise-grade implementations."
},
{
"role": "user",
"content": "Implement an enterprise-grade Model Context Protocol (MCP) server class 'FastMcpEngine_000' i... |
p16_sft_d01_0001 | 1 | Model Context Protocol (MCP) Server Protocol | [{"type": "function", "function": {"name": "mcp_server_dispatch_1", "description": "Executes standard MCP JSON-RPC 2.0 tool invocation with schema validation and capability discovery.", "parameters": {"type": "object", "properties": {"server_uri": {"type": "string"}, "method": {"type": "string", "enum": ["tools/call", ... | Implement an enterprise-grade Model Context Protocol (MCP) server class 'ModelContextProtocolHub_001' in Python complying with JSON-RPC 2.0 standards. It must validate tool schemas, enforce session token authentication, apply resilience bounds, and output structured execution telemetry. | <think>
1. Specification Decomposition & Architectural Analysis:
- Target Domain: Model Context Protocol (MCP) Server Protocol enterprise implementation.
- Primary Execution Target: Class 'ModelContextProtocolHub_001', Core Method: 'dispatch_tool_call'.
- Selected Data Modeling Strategy: Schema Paradigm #1.
... | import time
import json
import logging
import secrets
import hashlib
import hmac
from typing import Dict, Any, List, Optional, Tuple
from dataclasses import dataclass, field
logger = logging.getLogger("ModelContextProtocolHub_001")
class McpProtocolException(Exception):
def __init__(self, message: str = "Model Co... | 139 | [
{
"role": "system",
"content": "You are an expert autonomous software engineer and tool-use agent capable of writing mission-critical, enterprise-grade implementations."
},
{
"role": "user",
"content": "Implement an enterprise-grade Model Context Protocol (MCP) server class 'ModelContextProtocol... |
p16_sft_d01_0002 | 1 | Model Context Protocol (MCP) Server Protocol | [{"type": "function", "function": {"name": "mcp_server_dispatch_2", "description": "Executes standard MCP JSON-RPC 2.0 tool invocation with schema validation and capability discovery.", "parameters": {"type": "object", "properties": {"server_uri": {"type": "string"}, "method": {"type": "string", "enum": ["tools/call", ... | Implement an enterprise-grade Model Context Protocol (MCP) server class 'McpAgentBridge_002' in Python complying with JSON-RPC 2.0 standards. It must validate tool schemas, enforce session token authentication, apply resilience bounds, and output structured execution telemetry. | <think>
1. Specification Decomposition & Architectural Analysis:
- Target Domain: Model Context Protocol (MCP) Server Protocol enterprise implementation.
- Primary Execution Target: Class 'McpAgentBridge_002', Core Method: 'dispatch_tool_call'.
- Selected Data Modeling Strategy: Schema Paradigm #2.
- Standa... | import time
import json
import logging
import secrets
import hashlib
import hmac
from typing import Dict, Any, List, Optional, Tuple, TypedDict, TypeGuard
logger = logging.getLogger("McpAgentBridge_002")
class McpProtocolException(Exception):
def __init__(self, message: str = "Model Context Protocol (MCP) Server ... | 137 | [
{
"role": "system",
"content": "You are an expert autonomous software engineer and tool-use agent capable of writing mission-critical, enterprise-grade implementations."
},
{
"role": "user",
"content": "Implement an enterprise-grade Model Context Protocol (MCP) server class 'McpAgentBridge_002' ... |
p16_sft_d01_0003 | 1 | Model Context Protocol (MCP) Server Protocol | [{"type": "function", "function": {"name": "mcp_server_dispatch_3", "description": "Executes standard MCP JSON-RPC 2.0 tool invocation with schema validation and capability discovery.", "parameters": {"type": "object", "properties": {"server_uri": {"type": "string"}, "method": {"type": "string", "enum": ["tools/call", ... | Implement an enterprise-grade Model Context Protocol (MCP) server class 'AutonomousToolRuntime_003' in Python complying with JSON-RPC 2.0 standards. It must validate tool schemas, enforce session token authentication, apply resilience bounds, and output structured execution telemetry. | <think>
1. Specification Decomposition & Architectural Analysis:
- Target Domain: Model Context Protocol (MCP) Server Protocol enterprise implementation.
- Primary Execution Target: Class 'AutonomousToolRuntime_003', Core Method: 'dispatch_tool_call'.
- Selected Data Modeling Strategy: Schema Paradigm #3.
-... | import time
import json
import logging
import secrets
import hashlib
import hmac
from typing import Dict, Any, List, Optional, Tuple, NamedTuple
logger = logging.getLogger("AutonomousToolRuntime_003")
class McpProtocolException(Exception):
def __init__(self, message: str = "Model Context Protocol (MCP) Server Pro... | 137 | [
{
"role": "system",
"content": "You are an expert autonomous software engineer and tool-use agent capable of writing mission-critical, enterprise-grade implementations."
},
{
"role": "user",
"content": "Implement an enterprise-grade Model Context Protocol (MCP) server class 'AutonomousToolRuntim... |
p16_sft_d01_0004 | 1 | Model Context Protocol (MCP) Server Protocol | [{"type": "function", "function": {"name": "mcp_server_dispatch_4", "description": "Executes standard MCP JSON-RPC 2.0 tool invocation with schema validation and capability discovery.", "parameters": {"type": "object", "properties": {"server_uri": {"type": "string"}, "method": {"type": "string", "enum": ["tools/call", ... | Implement an enterprise-grade Model Context Protocol (MCP) server class 'EnterpriseMcpGateway_004' in Python complying with JSON-RPC 2.0 standards. It must validate tool schemas, enforce session token authentication, apply resilience bounds, and output structured execution telemetry. | <think>
1. Specification Decomposition & Architectural Analysis:
- Target Domain: Model Context Protocol (MCP) Server Protocol enterprise implementation.
- Primary Execution Target: Class 'EnterpriseMcpGateway_004', Core Method: 'dispatch_tool_call'.
- Selected Data Modeling Strategy: Schema Paradigm #0.
- ... | import time
import json
import logging
import secrets
import hashlib
import hmac
from typing import Dict, Any, List, Optional, Tuple
from pydantic import BaseModel, Field, ConfigDict
logger = logging.getLogger("EnterpriseMcpGateway_004")
class McpProtocolException(Exception):
def __init__(self, message: str = "Mo... | 125 | [
{
"role": "system",
"content": "You are an expert autonomous software engineer and tool-use agent capable of writing mission-critical, enterprise-grade implementations."
},
{
"role": "user",
"content": "Implement an enterprise-grade Model Context Protocol (MCP) server class 'EnterpriseMcpGateway... |
p16_sft_d01_0005 | 1 | Model Context Protocol (MCP) Server Protocol | [{"type": "function", "function": {"name": "mcp_server_dispatch_5", "description": "Executes standard MCP JSON-RPC 2.0 tool invocation with schema validation and capability discovery.", "parameters": {"type": "object", "properties": {"server_uri": {"type": "string"}, "method": {"type": "string", "enum": ["tools/call", ... | Implement an enterprise-grade Model Context Protocol (MCP) server class 'FastMcpEngine_005' in Python complying with JSON-RPC 2.0 standards. It must validate tool schemas, enforce session token authentication, apply resilience bounds, and output structured execution telemetry. | <think>
1. Specification Decomposition & Architectural Analysis:
- Target Domain: Model Context Protocol (MCP) Server Protocol enterprise implementation.
- Primary Execution Target: Class 'FastMcpEngine_005', Core Method: 'dispatch_tool_call'.
- Selected Data Modeling Strategy: Schema Paradigm #1.
- Standar... | import time
import json
import logging
import secrets
import hashlib
import hmac
from typing import Dict, Any, List, Optional, Tuple
from dataclasses import dataclass, field
logger = logging.getLogger("FastMcpEngine_005")
class McpProtocolException(Exception):
def __init__(self, message: str = "Model Context Prot... | 133 | [
{
"role": "system",
"content": "You are an expert autonomous software engineer and tool-use agent capable of writing mission-critical, enterprise-grade implementations."
},
{
"role": "user",
"content": "Implement an enterprise-grade Model Context Protocol (MCP) server class 'FastMcpEngine_005' i... |
p16_sft_d01_0006 | 1 | Model Context Protocol (MCP) Server Protocol | [{"type": "function", "function": {"name": "mcp_server_dispatch_6", "description": "Executes standard MCP JSON-RPC 2.0 tool invocation with schema validation and capability discovery.", "parameters": {"type": "object", "properties": {"server_uri": {"type": "string"}, "method": {"type": "string", "enum": ["tools/call", ... | Implement an enterprise-grade Model Context Protocol (MCP) server class 'ModelContextProtocolHub_006' in Python complying with JSON-RPC 2.0 standards. It must validate tool schemas, enforce session token authentication, apply resilience bounds, and output structured execution telemetry. | <think>
1. Specification Decomposition & Architectural Analysis:
- Target Domain: Model Context Protocol (MCP) Server Protocol enterprise implementation.
- Primary Execution Target: Class 'ModelContextProtocolHub_006', Core Method: 'dispatch_tool_call'.
- Selected Data Modeling Strategy: Schema Paradigm #2.
... | import time
import json
import logging
import secrets
import hashlib
import hmac
from typing import Dict, Any, List, Optional, Tuple, TypedDict, TypeGuard
logger = logging.getLogger("ModelContextProtocolHub_006")
class McpProtocolException(Exception):
def __init__(self, message: str = "Model Context Protocol (MCP... | 131 | [
{
"role": "system",
"content": "You are an expert autonomous software engineer and tool-use agent capable of writing mission-critical, enterprise-grade implementations."
},
{
"role": "user",
"content": "Implement an enterprise-grade Model Context Protocol (MCP) server class 'ModelContextProtocol... |
p16_sft_d01_0007 | 1 | Model Context Protocol (MCP) Server Protocol | [{"type": "function", "function": {"name": "mcp_server_dispatch_7", "description": "Executes standard MCP JSON-RPC 2.0 tool invocation with schema validation and capability discovery.", "parameters": {"type": "object", "properties": {"server_uri": {"type": "string"}, "method": {"type": "string", "enum": ["tools/call", ... | Implement an enterprise-grade Model Context Protocol (MCP) server class 'McpAgentBridge_007' in Python complying with JSON-RPC 2.0 standards. It must validate tool schemas, enforce session token authentication, apply resilience bounds, and output structured execution telemetry. | <think>
1. Specification Decomposition & Architectural Analysis:
- Target Domain: Model Context Protocol (MCP) Server Protocol enterprise implementation.
- Primary Execution Target: Class 'McpAgentBridge_007', Core Method: 'dispatch_tool_call'.
- Selected Data Modeling Strategy: Schema Paradigm #3.
- Standa... | import time
import json
import logging
import secrets
import hashlib
import hmac
from typing import Dict, Any, List, Optional, Tuple, NamedTuple
logger = logging.getLogger("McpAgentBridge_007")
class McpProtocolException(Exception):
def __init__(self, message: str = "Model Context Protocol (MCP) Server Protocol e... | 131 | [
{
"role": "system",
"content": "You are an expert autonomous software engineer and tool-use agent capable of writing mission-critical, enterprise-grade implementations."
},
{
"role": "user",
"content": "Implement an enterprise-grade Model Context Protocol (MCP) server class 'McpAgentBridge_007' ... |
p16_sft_d01_0008 | 1 | Model Context Protocol (MCP) Server Protocol | "[{\"type\": \"function\", \"function\": {\"name\": \"mcp_server_dispatch_8\", \"description\": \"Ex(...TRUNCATED) | "Implement an enterprise-grade Model Context Protocol (MCP) server class 'AutonomousToolRuntime_008'(...TRUNCATED) | "<think>\n1. Specification Decomposition & Architectural Analysis:\n - Target Domain: Model Contex(...TRUNCATED) | "import time\nimport json\nimport logging\nimport secrets\nimport hashlib\nimport hmac\nfrom typing (...TRUNCATED) | 126 | [{"role":"system","content":"You are an expert autonomous software engineer and tool-use agent capab(...TRUNCATED) |
p16_sft_d01_0009 | 1 | Model Context Protocol (MCP) Server Protocol | "[{\"type\": \"function\", \"function\": {\"name\": \"mcp_server_dispatch_9\", \"description\": \"Ex(...TRUNCATED) | "Implement an enterprise-grade Model Context Protocol (MCP) server class 'EnterpriseMcpGateway_009' (...TRUNCATED) | "<think>\n1. Specification Decomposition & Architectural Analysis:\n - Target Domain: Model Contex(...TRUNCATED) | "import time\nimport json\nimport logging\nimport secrets\nimport hashlib\nimport hmac\nfrom typing (...TRUNCATED) | 134 | [{"role":"system","content":"You are an expert autonomous software engineer and tool-use agent capab(...TRUNCATED) |
End of preview. Expand in Data Studio
β‘ Agentic Tool-Use & Function Calling Suite (2026 Edition)
π The Definitive 2026 Training Suite for Function Calling, Model Context Protocol (MCP), and Autonomous Software Agents.
π Dataset Overview
Standard open-source function-calling datasets are saturated with 10-line toy stubs, unhandled exceptions, and naive wrappers that cause models to crash under real production conditions.
The Agentic Tool-Use & Function Calling Suite (2026) enforces a Heavyweight Enterprise Tier-1 Standard:
- 126.8 Dense LOC Average (Range 115β142 LOC): Every chosen implementation features Pydantic V2 schemas, frozen dataclass slots, Circuit Breakers, Token Buckets, OTel spans, and Prometheus metrics.
- 100.0% Real AST Entropy: Generated via an orthogonal 5-axis combinatorial synthesis engine (1,280 structural schemas per domain). Zero repetitive skeletons.
- Zero-Tolerance Stubs Policy: Exactly 0 instances of
TODO,FIXME,STUB,placeholder, orpassacross 100% of rows. - Deep Multi-Phase CoT Reasoning: 29 lines of structured
<think>trace covering threat modeling, SLA latency bounds, injection defenses, and state rollbacks. - Native ChatML (
messages) Schema: Plug-and-play compatible with Unsloth, LLaMA-Factory, Axolotl, and TRL out of the box.
π 20 Mission-Critical Enterprise Domains
| ID | Domain | Architecture Highlights |
|---|---|---|
| 01 | Model Context Protocol (MCP) | JSON-RPC 2.0 stdio server, tool discovery, parameter schema validation |
| 02 | OpenAPI 3.1 / Swagger Dispatcher | REST routing, HTTP verbs, and path traversal security guards |
| 03 | Multi-Turn ReAct Agent Engine | Reasoning-action-observation cycles, step horizons, and scratchpad history |
| 04 | SQL & Database Query Tool | Read-only SELECT enforcement, DDL/DML mutation bans, and row limits |
| 05 | Shell & Sandboxed CLI Guard | POSIX argv execution, binary whitelisting, and metacharacter sanitization |
| 06 | File System & Workspace Mutation | Chroot sandbox jail, path-traversal blocking, and SHA256 atomic commits |
| 07 | Git & Version Control Tooling | Refspec validation, atomic commits, and GPG signature verification |
| 08 | Web Search & SERP Extraction | Query parameter escaping, domain deduplication, and relevance scoring |
| 09 | Web Scraping & DOM Traversal | SSRF cloud metadata (169.254.169.254) defense and script stripping |
| 10 | Vector Search & RAG Dispatcher | HNSW KNN similarity, embedding dimension alignment, and score thresholds |
| 11 | Graph Database & Cypher Engine | openCypher pattern matching, hop bounds, and mutation prevention |
| 12 | Workflow Orchestrator (DAG Engine) | Topological sorting, cycle detection (Kahn DFS), and parallel scheduling |
| 13 | Memory & State Persister Agent | Tenant session isolation, AES-256 GCM encryption, and TTL pruning |
| 14 | Dynamic Code Interpreter (REPL) | AST import inspection, forbidden syscall filtering, and memory quotas |
| 15 | Computer Vision & Multi-Modal Tool | MIME validation, decompression bomb defense, and bbox coordinates |
| 16 | Audio & Speech STT/TTS Processor | Sample rate verification (16kHz/44.1kHz), codec validation, and stream bounds |
| 17 | Financial Market Data & Order Router | FIX 5.0 SP2 wire format, tick sizes, and SEC 15c3-5 pre-trade risk collars |
| 18 | IoT & Hardware Telemetry Client | MQTT v5.0 ingestion, timestamp skew rejection, and physical metric bounds |
| 19 | Cybersecurity Vulnerability & CVE Scanner | SemVer tuple parsing, OSV/NVD feeds, and CVSS v3.1 severity tiers |
| 20 | Self-Reflection & Plan-Correction | Execution loop detection, goal divergence tracking, and corrective backtracking |
π» Quickstart (Fine-Tuning with Unsloth / LLaMA-Factory)
from datasets import load_dataset
# Load preview dataset directly from Hugging Face
dataset = load_dataset("antigravity/agentic-tool-use-suite-2026", split="train")
print(f"Loaded {len(dataset)} samples across {len(set(dataset['domain']))} domains!")
print(f"Sample messages: {dataset[0]['messages']}")
π Get the Full 12,500-Row Commercial Suite (10k SFT + 2.5k DPO)
Need the full dataset for training frontier models?
- π¦ Complete 10,000 SFT Records + 2,500 DPO Pairs
- ποΈ Indexed SQLite Database (
product16_agentic_suite.db, 170 MB) - π Commercial B2B Deployment License
- π Available on Gumroad ($9 / $29 / $59): https://gumroad.com/l/agentic-tool-use-suite-2026
- Downloads last month
- 32