Tuning Engines CLI & MCP Server

Govern every AI workflow through one API.

Tuning Engines is a governed AI runtime for model, agent, skill, and MCP workflows. Route inference through one OpenAI-compatible API, apply RBAC and traffic policies, request approvals for high-risk actions, inspect traces and usage, and connect durable orchestration frameworks such as LangGraph and Temporal. The same CLI and MCP server also manage domain-specific fine-tuning of open-source models.

Training Agents

Tuning Engines uses specialized agents that control how your data is analyzed and converted into training data. Each agent produces a different kind of domain-specific fine-tuned model optimized for its use case. Current agents focus on code, with more coming for customer support, data extraction, security review, ops, and other domains.

Cody (code_repo) — Code Autocomplete Agent

Cody fine-tunes on your GitHub repo using QLoRA (4-bit quantized LoRA) via the Axolotl framework (HuggingFace Transformers + PEFT). It learns your codebase's patterns, naming conventions, and project structure to produce a fast, lightweight adapter optimized for real-time completions.

Best for: code autocomplete, inline suggestions, tab-complete, code style matching, pattern completion.

te jobs create --agent code_repo \
  --base-model Qwen/Qwen2.5-Coder-7B-Instruct \
  --repo-url https://github.com/your-org/your-repo \
  --output-name my-cody-model

SIERA (sera_code_repo) — Bug-Fix Specialist

SIERA (Synthetic Intelligent Error Resolution Agent) uses the Open Coding Agents approach from AllenAI to generate targeted bug-fix training data from your repository. It synthesizes realistic error scenarios and their resolutions, then fine-tunes a model that learns your team's debugging style, error handling conventions, and fix patterns.

Best for: debugging, error resolution, patch generation, root cause analysis, fix suggestions.

te jobs create --agent sera_code_repo \
  --quality-tier high \
  --base-model Qwen/Qwen2.5-Coder-7B-Instruct \
  --repo-url https://github.com/your-org/your-repo \
  --output-name my-siera-model

Quality tiers (SIERA only):

• low — Faster, fewer synthetic pairs (default)
• high — Deeper analysis, more training data, better results

Coming Soon

Agent	Persona	What it does
Resolve	Mira	Fine-tunes on support tickets, macros, and KB articles for automated ticket resolution
Extractor	Flux	Trains for strict schema extraction from docs, PDFs, and business text
Guard	Aegis	Security-focused code reviewer that catches risky patterns and proposes safer fixes
OpsPilot	Atlas	Incident response agent trained on runbooks, postmortems, and on-call notes

Supported Base Models

Size	Models
3B	Qwen/Qwen2.5-Coder-3B-Instruct
7B	codellama/CodeLlama-7b-hf, deepseek-ai/deepseek-coder-7b-instruct-v1.5, Qwen/Qwen2.5-Coder-7B-Instruct
13-15B	codellama/CodeLlama-13b-Instruct-hf, bigcode/starcoder2-15b, Qwen/Qwen2.5-Coder-14B-Instruct
32-34B	deepseek-ai/deepseek-coder-33b-instruct, codellama/CodeLlama-34b-Instruct-hf, Qwen/Qwen2.5-Coder-32B-Instruct
70-72B	codellama/CodeLlama-70b-Instruct-hf, meta-llama/Llama-3.1-70B-Instruct, Qwen/Qwen2.5-72B-Instruct

Quick Start

npm install -g tuningengines-cli

# Or run without installing
npx -y --package tuningengines-cli@latest te auth status

# Sign up or log in (opens browser — works for new accounts too)
te auth login

# Add credits (opens browser to billing page)
te billing add-credits

# Estimate cost before training
te jobs estimate --base-model Qwen/Qwen2.5-Coder-7B-Instruct

# Train Cody on your repo
te jobs create --agent code_repo \
  --base-model Qwen/Qwen2.5-Coder-7B-Instruct \
  --repo-url https://github.com/your-org/your-repo \
  --output-name my-model

# Monitor training
te jobs status <job-id> --watch

# View your trained models
te models list

# Create a governed orchestration starter
te orchestration init langgraph
te orchestration init temporal
te orchestration init inngest
te orchestration init triggerdev
te orchestration init hatchet
te orchestration init restate
te orchestration init dbos
te orchestration init dapr
te orchestration init prefect
te orchestration init dagster
te orchestration init airflow

MCP Server Setup

The CLI includes a built-in MCP server with 60+ tools. Any AI assistant that supports MCP can fine-tune models, manage training jobs, run evaluations, check inference usage, inspect traces, review approvals, and manage non-secret tenant registry metadata through natural language.

For security, the MCP server intentionally does not expose internal proxy routes. It also refuses MCP-side inference-key creation and raw secret-bearing mutation fields. Use the CLI or web UI for workflows that intentionally create one-time keys, submit raw provider secrets, validate S3 credentials, or import/export S3 assets with raw credentials.

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "tuning-engines": {
      "command": "npx",
      "args": ["-y", "--package", "tuningengines-cli@latest", "te", "mcp", "serve"],
      "env": {
        "TE_API_KEY": "te_your_key_here"
      }
    }
  }
}

Claude Code

claude mcp add tuning-engines -- npx -y --package tuningengines-cli@latest te mcp serve

Work Sessions and outcomes

Label the desired outcome for a project without interrupting your coding workflow:

te goal start "Fix flaky checkout retries"
te goal show
te goal complete --result succeeded

Install optional native telemetry hooks for Claude Code or Codex:

te guard claude-code install --mode observe --project .
te guard claude-code doctor
te guard claude-code doctor --probe
te guard codex install
te guard codex doctor

Claude Code writes project-local hooks into .claude/settings.local.json. On
Windows, verify with dir .\.claude, type .\.claude\settings.local.json,
then restart Claude Code from the same project root and review claude /hooks.
doctor --probe is available in tuningengines-cli 0.4.20 and later; it runs
synthetic hook events through the installed commands and checks that the trace is
visible to Tuning Engines. Hook invocations also write a local redacted status
log at .claude/tuning-engines-hook-status.jsonl.
Codex project hooks require review and trust from /hooks. The installer pins
hooks to the exact CLI executable that created them; te guard codex doctor
reports that path, CLI version, native-event contract, required lifecycle
hooks, and any conflicting te command found on PATH. Each Codex or Claude
Code user prompt starts a new trace while retaining one stable Work Session for
the native conversation. Tuning Engines sends pseudonymous session and
transcript references by default, not transcript contents or local absolute
paths.

Claude Code Plugin

The repository also ships a Claude Code plugin wrapper around the same MCP
server. It keeps installation discoverable while preserving the same
TE_API_KEY environment-variable boundary:

claude plugin marketplace add cerebrixos-org/tuning-engines-cli
claude plugin install tuning-engines@tuning-engines

The same plugin directory includes a Codex plugin manifest, marketplace entry,
MCP server, and tenant-operations skill. It uses the Tuning Engines web app as
the shared control-plane UI for traces, approvals, policies, cost, and review.

DeepSeek Harness Plugin

Install the native Harness bundle to capture turn, model-step, and tool
lifecycle telemetry and optionally enforce TE governance before tool execution:

export TE_API_KEY="your-tenant-token-or-inference-key"
dsh plugin --profile default add tuningengines-cli

The adapter uses a disk-backed local spool, background batching, bounded
retries, stable tool-call correlation, and metadata-only capture by default.

VS Code / Cursor / Windsurf

Add to your MCP settings (.vscode/mcp.json or equivalent):

{
  "servers": {
    "tuning-engines": {
      "command": "npx",
      "args": ["-y", "--package", "tuningengines-cli@latest", "te", "mcp", "serve"],
      "env": {
        "TE_API_KEY": "te_your_key_here"
      }
    }
  }
}

What the AI assistant can do

When connected, your AI assistant can:

• "Fine-tune Qwen 7B on my-org/my-repo using the SIERA agent with high quality"
• "How much would it cost to train a 32B model for 3 epochs on this repo?"
• "Check the status of my latest training job"
• "List my trained models"
• "Export my model to s3://my-bucket/models/"
• "Show my account balance"
• "Train a bug-fix specialist on this repo" (auto-selects SIERA)
• "Create an autocomplete model for this codebase" (auto-selects Cody)

The create_job tool description includes full agent details and model lists, so AI assistants automatically select the right agent and model based on what you ask for.

Unified API Endpoint

Tuning Engines can be used anywhere a tool accepts an OpenAI-compatible API
base URL. Point the client at:

https://api.tuningengines.com/v1

Use an inference key that starts with sk-te-... for live model calls, and use
the model IDs shown by:

te inference models

This lets OpenCode, Temporal activities, LangGraph apps, OpenAI SDK clients,
and other custom-provider clients route through the same Tuning Engines control
plane for model RBAC, routing, fallbacks, guardrails, AGT policy, traces,
usage metering, and cost attribution.

See docs/unified-api-endpoint.md for copy-paste
examples for OpenCode, Temporal, Python, JavaScript, and other
OpenAI-compatible clients.

Agent Runtime SDK and Orchestration Starters

Use the CLI/MCP package when you want npx tools for assistants. Use the
Python SDK when you want your own app to run durable agent workflows while
Tuning Engines remains the governed control plane for models, agents, skills,
MCP tools, RBAC, AGT policy, audit, usage, and token economics.

OpenAI Agents SDK users can install the native model and trace integration:

pip install "tuning-engines[openai-agents]"

See packages/tuning-agents/README.md
for setup. The OpenAI Agents runtime owns the loop; TE supplies the governed
endpoint and receives metadata-only SDK traces through its background processor.

Install the published Python SDK:

pip install "tuning-engines[langgraph]"
pip install "tuning-engines[temporal]"

The package installs the tuning_agents Python module.

LangGraph example:

from langgraph.checkpoint.memory import InMemorySaver

from tuning_agents import TuningClient
from tuning_agents.langgraph import create_tuning_langgraph_agent, invoke_with_trace

client = TuningClient(api_key="te_your_key_here")

agent = create_tuning_langgraph_agent(
    client,
    model="llama-3.3-70b-fp8",
    agent_names=["billing-escalation"],
    checkpointer=InMemorySaver(),
    interrupt_before=["tools"],
)

result = invoke_with_trace(
    client,
    agent,
    [{"role": "user", "content": "Triage this ticket and escalate if needed."}],
    thread_id="ticket-123",
)

client.flush_trace(name="ticket-triage", runtime="langgraph", status="succeeded")

Temporal example:

from tuning_agents.temporal import (
    TuningEnginesTemporalFeatures,
    create_tuning_engines_plugin,
    define_temporal_workflow,
)

plugin = create_tuning_engines_plugin(
    features=TuningEnginesTemporalFeatures(
        built_in_workflow=False,
        model_calls=True,
        skill_tools=True,
        mcp_tools=True,
        agents=True,
        approvals=True,
        traces=True,
        state_references=True,
        interventions=True,
        model_catalog=True,
        usage=True,
    )
)
TuningAgentWorkflow = define_temporal_workflow()
# Pass plugin to Client.connect(..., plugins=[plugin]) and register the workflow.

The SDK captures runtime events from LangGraph/Temporal and posts them to
POST /api/v1/traces. Each event carries a run_id, request_id, and a
normalized event type such as model.call, mcp.tool_call, agent.message,
workflow.step, human.edit, action.finalized, outcome.recorded, or
state.reference. The app pairs that with inference usage, request capture,
policy decisions, approval requests, external state references, audit, and
billing logs.

JavaScript/TypeScript users can also import lightweight tracing helpers from
the npm package:

import { createOpenAIAgentsTraceAdapter } from "tuningengines-cli/adapters/openai-agents";
import { createClaudeAgentSdkTraceAdapter } from "tuningengines-cli/adapters/claude-agent-sdk";

Both helpers send redacted run, model, tool, handoff, error, goal, and outcome
events to the existing trace API. goal_key, goal_status, and goal_score
are normalized into the same success-signal analytics as outcome_key.

For decision traces, store redacted signals in metadata.decision, for example
proposal_summary, changed_fields, change_summary, final_action,
outcome_label, and reason_summary. Do not place raw prompts, provider keys,
tenant secrets, or full customer data in trace metadata.

Generate a starter kit:

te orchestration init langgraph --dir ./lg-te-demo
te orchestration init temporal --dir ./temporal-te-demo
te orchestration init inngest --dir ./inngest-te-demo
te orchestration init triggerdev --dir ./trigger-te-demo
te orchestration init hatchet --dir ./hatchet-te-demo
te orchestration init restate --dir ./restate-te-demo
te orchestration init dbos --dir ./dbos-te-demo
te orchestration init dapr --dir ./dapr-te-demo
te orchestration init prefect --dir ./prefect-te-demo
te orchestration init dagster --dir ./dagster-te-demo
te orchestration init airflow --dir ./airflow-te-demo

LangGraph and Temporal starters use the Python runtime SDK. Inngest,
Trigger.dev, and Hatchet starters generate TypeScript projects with a small
self-contained Tuning Engines helper. Restate, DBOS, and Dapr starters use the
same TypeScript helper. Prefect, Dagster, and Airflow starters generate Python
workflow examples with a small helper module. All generated examples include
governed model calls, trace flushing, registry manifests, policy context
metadata, decision metadata, runtime state references, and approval retry
patterns.

CLI Commands

Authentication

Command	Description
te auth login	Sign up or log in via browser
te auth logout	Clear saved credentials
te auth status	Show current auth status (email, balance)

Training Jobs

Command	Description
te jobs list	List all training jobs
te jobs show <id>	Show job details
te jobs create	Submit a training job (--agent, --quality-tier, --base-model, --repo-url, --output-name)
te jobs status <id>	Live status (--watch for continuous polling)
te jobs cancel <id>	Cancel a running job
te jobs retry <id>	Retry from last checkpoint
te jobs estimate	Cost estimate before submitting
te jobs validate-s3	Pre-validate S3 credentials

Models

Command	Description
te models list	List your trained models
te models show <id>	Show model details
te models base	List supported base models
te models import	Import a model from S3
te models export <id>	Export a model to S3
te models delete <id>	Delete a model
te models status <id>	Check import/export status

Datasets

Command	Description
te datasets list	List all datasets
te datasets show <id>	Show dataset details
te datasets create	Create a dataset from S3 (--name, --s3-url, --for-evaluation)
te datasets delete <id>	Delete a dataset
te datasets status <id>	Check import/processing status

Evaluations

Command	Description
te evals list	List all evaluations
te evals show <id>	Show evaluation details and scores
te evals create	Run an evaluation (--model, --dataset, --evaluators)
te evals cancel <id>	Cancel a running evaluation
te evals status <id>	Live evaluation progress
te evals evaluators	List available evaluators
te evals estimate	Cost estimate for an evaluation

Inference

Command	Description
te inference models	List available inference models
te inference usage	Show inference API usage stats
te inference jwt	Get a JWT for direct API access
te inference token	Exchange an inference key (sk-te-...) for a short-lived inference JWT
te inference chat --data '<json>'	Run an OpenAI-compatible chat completion
te inference responses --data '<json>'	Run an OpenAI Responses request
te inference embeddings --data '<json>'	Create embeddings
te inference messages --data '<json>'	Run an Anthropic-compatible Messages request

Runtime Traces and Approvals

Command	Description
te traces list	List LangGraph, Temporal, and custom runtime traces
te traces show <run-id>	Show one trace, including events, policy decisions, and approvals when linked
te traces ingest --data '<json>'	Ingest or update a trace using a user API token or inference key
te outcomes list	List observed outcomes, goals, evals, and workflow success signals
te outcomes record --run-id ... --key ... --label ...	Record a success signal for a run
te outcomes map --outcome-key ... --criteria '<json>'	Map unmapped events to an outcome key
te insights list	List Insight Loop recommendations
te insights accept <id>	Accept an insight as valid; does not change production
te insights apply <id>	Apply or queue the approved action for an accepted insight
te doctor simulate --data '<json>'	Simulate inference access, role, endpoint, policy, and resource checks
te policy-decisions list	List AGT YAML policy decisions
te policy-decisions show <id>	Show one policy decision with redacted context
te policy-templates list	List curated AGT YAML policy templates
te policy-templates render <id> --params '<json>'	Render disabled/shadow policy YAML from safe structured parameters
te policy-drafts generate --prompt '<text>'	Generate an AI-assisted disabled/shadow draft for review and testing
te approvals list --status pending	List policy approval requests
te approvals show <id>	Show approval detail and retry metadata
te approvals approve <id>	Approve a pending request
te approvals deny <id>	Deny a pending request
te interventions list/show/request/ack/complete/fail	Manage runtime pause, resume, cancel, and replay requests
te state list/show/upsert	Manage safe external workflow-state and memory references
te registry sync --file tuning-registry.yml --dry-run/--apply	Diff or apply agent, skill, and MCP registry manifests
te registry show <id>	Inspect an applied registry sync
te work-sessions list/show/complete/confirm-outcome	Inspect and update Work Sessions
te initiatives list/show/create/update	Manage strategic initiative groupings

Orchestration Starters

Command	Description
te orchestration init langgraph	Create a LangGraph starter wired to Tuning Engines governance and traces
te orchestration init temporal	Create a Temporal worker starter wired to Tuning Engines governance and traces
te orchestration init inngest	Create an Inngest function starter wired to Tuning Engines governance and traces
te orchestration init triggerdev	Create a Trigger.dev task starter wired to Tuning Engines governance and traces
te orchestration init hatchet	Create a Hatchet workflow starter wired to Tuning Engines governance and traces
te orchestration init restate	Create a Restate service starter wired to Tuning Engines governance and traces
te orchestration init dbos	Create a DBOS workflow starter wired to Tuning Engines governance and traces
te orchestration init dapr	Create a Dapr Workflow starter wired to Tuning Engines governance and traces
te orchestration init prefect	Create a Prefect flow starter wired to Tuning Engines governance and traces
te orchestration init dagster	Create a Dagster asset starter wired to Tuning Engines governance and traces
te orchestration init airflow	Create an Airflow DAG starter wired to Tuning Engines governance and traces

Agents

Command	Description
te agents list	List available agents
te agents show <id>	Show agent details and capabilities
te agents message <name> --data '<json>'	Send a governed A2A agent message

Skills and MCP Execution

Command	Description
te skills list	List skills visible to the inference identity
te skills prepare <name> --data '<json>'	Prepare a governed skill
te skills invoke <name> --data '<json>'	Invoke a governed skill
te mcp call --server <name> --tool <name> --arguments '<json>'	Call an enabled governed MCP tool
te mcp rediscover <server-id>	Refresh MCP tool discovery
te mcp tools list/enable/disable/toggle	Administer discovered MCP tools

Compliance Automation

Command	Description
te compliance validate --data '<json>'	Validate bounded content against adopted rulepacks
te compliance rewrite --data '<json>'	Produce and revalidate a safe rewrite
te compliance evidence <id>	Inspect compliance evidence
te compliance risks list/show/create/update	Manage the tenant risk register
te compliance risks assess/map-control/add-subject/remove-subject	Operate the risk-to-control lifecycle
te compliance source-runs create/show/submit-results/complete	Ingest normalized scanner or webhook results
te compliance certifications create/show	Run and inspect compliance certification jobs

Tenant Admin Automation

These commands require an API token for a tenant owner or tenant admin. They are
designed for CI smoke tests and end-to-end product checks. Secret fields can be
sent on create/update where the server supports them, but responses never print
stored provider keys, AWS secrets, or invitation tokens.

Command	Description
te tenant resources	List supported tenant resource names
te tenant list <resource>	List resources such as inference_keys, inference_roles, model_deployments, routing_profiles, guardrail_policies, governance_policies, mcp_servers, tenant_agents, tenant_skills, and credential_sources
te tenant show <resource> <id>	Show one tenant resource
te tenant create <resource> --data '<json>'	Create a