This section introduces the DeepSeek Harness observability log application for large model monitoring, available in the TLS application marketplace.

Overview

DeepSeek Harness is an application framework developed by DeepSeek AI for building, running, and scaling agents, adopting an everything is a plugin architecture.

After the DeepSeek Harness observability log application is integrated, TLS automatically creates a dedicated Trace log topic for DeepSeek Harness, generates visual charts, and displays session, Trace, Token, and tool invocation data, enabling trace tracking and observability. It can be used in the following scenarios:

Observability	Description
Performance analysis	Model invocation latency distribution Tool execution time ranking Token consumption trend
Error tracking	View the complete invocation chain Identify the error stage Analyze the cause of the error
Cost optimization	Identify high-consumption conversations Optimize prompts to reduce tokens Select more cost-effective models

Prerequisites

Category	Description
Software limits	The server deploying DeepSeek Harness must have NPM and Node.js version 22 or later installed.
Region restriction	The server deploying DeepSeek Harness and the log service must be in the same region to ensure network connectivity.
Authentication method	Obtain the user's Access Key ID, Secret Access Key, or APIKey according to the authentication method. AK/SK: AccessKey ID and AccessKey Secret for Volcano Engine accounts. Obtain them from the Key Management page in the Volcano Engine console. We recommend using IAM user keys. Ensure this user has the required permissions. For details, see IAM overview. Supports automatic creation of TLS resources (project/app/topic) and auto-generates log topic IDs. API Key: A key credential for requesting Volcano TLS. It is used only for reporting Log, Metric, and Trace data. Enforces strict permissions and does not support auto creation of TLS resources. It is suitable for multi-data center environments and strict permission management scenarios. Notice API Key is in closed beta. If needed, contact your account manager for access to the allowlist.

1. Integrate the DeepSeek Harness log application

Log in to TLS console.

Select the region where log service is located from the top navigation bar.

In the left navigation bar, select AgentLoop > AI observability.

On the AI observability page, click Integrate AI observability.

On the Integrate AI observability page, create an application.

Click the DeepSeek Harness icon and select the DeepSeek Harness application.

Configure the integration application parameters, then click Create application.

Parameters	Description
Region	The region where log application data is stored.
Log project	Select the project where log application data is stored. Notice After the log application is successfully created, a log topic with the DeepSeek Harness tag will be automatically created in the project to store trace data. The relevant information for this log topic is as follows: Log topic name: application name_trace Number of shards: 2 Log storage period: Standard storage for 90 days Index: Key-value index is automatically configured.
Application name	Enter the name of the DeepSeek Harness log application. Length must be between 3 and 64 characters. Only supports Chinese, lowercase letters, numbers, underscores (_), and hyphens (-). Can not start or end with a hyphen (-) or underscore (_). Cannot start with tls_, tos_, or certain reserved fields.
Description	Log application description information. Length limit is 0–128 characters.

After creating the application, proceed to the next step to obtain usage instructions and integrate the log application.

2. Integrate DeepSeek Harness application data

Obtain usage instructions for integrating DeepSeek Harness application data.

Go to the integration configuration page.

Obtain when integrating the log application
Obtain after integrating the log application
When creating a DeepSeek Harness log application in AgentLoop, follow the interface prompts to obtain the corresponding usage guide.
Refer to 1. Integrate the DeepSeek Harness log application to integrate the DeepSeek Harness log application.
On the Integrate AI observability page, in Integration configuration, configure the following information to obtain the usage guide.
Log in to TLS console.
Select the region where log service is located from the top navigation bar.
In the left navigation bar, select AgentLoop > AI observation.
On the AI observation page, access the usage guide page using the following methods.
Click the access configuration icon in the Operation column of the target log application.
Click the target log application name, then click Access management.

On the integration configuration page, select the authentication method, configure the following information, and obtain usage instructions.

AK/SK access
API Key access
For Access method, select AK SK access.
For Connection method, select Public network or Volcano Engine intranet based on actual requirements.
In Usage guide, copy the Auto generation usage guide.
For Access method, select API Key access.
For Connection method, select Public network or Volcano Engine intranet based on actual requirements.
For API Key, select the API Key that has been created.
In Usage guide, copy the Auto generation usage guide.

Usage instruction example (using AK/SK integration):

Public network
Volcano Engine intranet

npm exec -y
--registry=https://registry.npmjs.org/
--package=@volcengine/tls-observer-dsh-install@latest --
tls-observer-dsh-install install
--profile web
--non-interactive
--force
--capture-content true
--region The system autopopulates the current region.
--trace-topic-id The system autopopulates the trace-topic-id
--ak <Requires manual replace of the actual tls-ak>
--sk <Requires manual replace of the actual tls-sk>

(AK/SK method only) Modify the authentication information in the usage instructions.

Enter the obtained Access Key ID and Secret Access Key into the AK and SK fields in the usage instructions, respectively.

Integrate DeepSeek Harness observability logs into the log application.

Log in to the server where DeepSeek Harness is deployed.

Use the information obtained from the usage instructions to install the TLS plugin.

After the plugin installation is complete, restart the DeepSeek Harness service.

Return to the TLS console to view the observability chart data.

Monitoring chart details

The monitoring chart details for the TLS DeepSeek Harness application observability are as follows:

Type	Chart
Homepage	Request overview: Number of sessions, conversation turns, model requests, tool invocation count, request trend Cost overview: Total Tokens, Input Tokens, Output Tokens, Cache Read Tokens, Reasoning Tokens, Token consumption trend, model invocation trend Tool overview Successful invocation count, failed invocation count, tool invocation latency P90, tool invocation Top 10, tool invocation failure Top 10, tool invocation trend Model performance overview Model invocation distribution Top 10, model invocation latency P90, provider invocation distribution, provider invocation latency P90
Trace analysis	Homepage Total trace count, total token consumption, average tokens, P90 duration Trace call chain analysis Displays the Trace ID, GenAIInput, GenAIOutput, status, total latency (ms), Total Tokens, Input Tokens, Output Tokens, start time, end time, and more for the specified call chain.
Session analysis	Homepage Session count, total token consumption, model request count, average session tokens Session analysis Displays the First Input, Last Output, Session ID, Total Tokens, call chain count, latency (s), model, start time, end time, and more for the specified session.
Token analysis	Homepage Total Tokens, Input Tokens, Output Tokens, Cache Read Tokens, Reasoning Tokens, Input Tokens (excluding Cache), Session Token consumption Top 10, Model Token consumption Top 10, Token category proportion
Topic list	Displays the log types, projects, log topics, description information, and more generated by the current log application.

Manage my applications

After the log application is successfully connected, you can manage connected log applications as follows:

AgentLoop

Application marketplace

Log in to TLS console.

Select the region where log service is located from the top navigation bar.

In the left navigation bar, select AgentLoop > AI Observation.

On the AI Observation page, in the Actions column for the corresponding log application, the following management actions are supported:

Edit application information: Click the edit icon to modify the application name and description.

Obtain usage guide: Click the access configuration icon to open the access configuration popup, where you can obtain example usage guide information.

Delete my application: Click the delete icon to remove the log application.

Application name_trace (log topic) log fields

Field	Type	Description
Attributes	json	Custom key-value pairs attached to the Span, recording business-related information.
Duration	long	Duration of the Span (End - Start), measuring the time consumed by the operation.
End	long	Span end time.
Events	text	Events that occur during the Span lifecycle.
Host	text	The host running the service.
Kind	text	Type of the Span.
Links	text	References to other Spans (used for cross-Trace associations, such as asynchronous task scenarios).
Name	text	Name of the Span, usually describing the operation.
OTLPName	text	Original name when reported via the OTLP protocol (OTLP = OpenTelemetry Protocol).
OTLPVersion	text	Version number of the OTLP protocol used.
ParentSpanID	text	ID of the parent Span, indicating which operation triggered the current operation. This field is empty for the root node.
Resource	json	A set of attributes describing the service itself (more comprehensive than ServiceName), such as service name, version, runtime environment, and so on.
ServiceName	text	Name of the service that generated this Span.
SpanID	text	Unique ID of the current Span.
Start	long	Span start time.
StatusCode	text	Execution result of the Span.
StatusDescription	text	Supplementary description of the execution result of the Span.
TracelD	text	TraceID of the entire call chain. All Spans for the same request share the same TraceID.
TraceState	text	Custom trace context metadata.

Attributes

Field	Type	Description
gen_ai.operation.name	text	Operation name: identifies the action type of this AI invocation.
gen_ai.conversation.id	text	Conversation ID: uniquely identifies a complete conversation.
gen_ai.agent.name	text	AI Agent name (OTel standard field): identifies the name of the agent that triggered this invocation.
gen_ai.input.messages	text	Input messages, the complete conversation history sent to the model.
gen_ai.output.messages	text	Output messages, the complete reply content returned by the model.
gen_ai.provider.name	text	Provider name.
gen_ai.request.model	text	Model specified in the request, the model name filled in by the client when initiating the request.
gen_ai.response.model	text	Actual response model, the model version actually used by the server.
gen_ai.tool.name	text	Tool name when the agent calls an external tool (Function Calling).
gen_ai.usage.input_tokens	long	Number of input tokens, total input tokens consumed by this request.
gen_ai.usage.output_tokens	long	Number of output tokens, total output tokens generated by this request.
gen_ai.usage.cache_read.input_tokens	long	Number of tokens read from cache, amount of tokens read from the prompt cache (cache hits can significantly reduce costs).
gen_ai.usage.cache_creation.input_tokens	long	Number of tokens written to cache, amount of tokens newly written to the prompt cache by this request.
gen_ai.usage.reasoning.output_tokens	long	Number of reasoning tokens, tokens consumed during the internal reasoning process of models such as DeepSeek-R1 (content within the <think> tag).
gen_ai.response.time_to_first_chunk	double	Time to first chunk (TTFC), elapsed time from sending the request to receiving the first streaming chunk, measures the speed at which the model begins to respond.
session.id	text	Session ID: associates multiple invocations with the same conversation session, used to reconstruct the complete multi-turn conversation chain.
error.type	text	Error type: records the error category when an invocation fails.
session.name	text	Session name.
tls.app.type	text	Application type of TLS log application.
tls.plugin.version	text	TLS plugin version.
agent.name	text	Agent name (TLS Custom field), same meaning as gen_ai.agent.name, Extra specs for the platform side.
gen_ai.agent.version	text	AI Agent version (OTel standard field): records the version number of the agent.
agent.version	text	Agent version (TLS Custom field), same meaning as gen_ai.agent.version.
user.name	text	User name, the user's readable name, used together with user.id.
user.id	text	User ID: identifies the end user who triggered this request, used for usage and cost statistics per user.

Resource

Field	Type	Description
tls.app.type	text	TLS application type, Volcano Engine TLS platform identifier, indicates which type of application or access method this log/trace comes from, such as deepseek_harness, llm, web, and so on, used for classification and filtering by application type in the observability platform
dsh.observer.version	text	The DeepSeek Harness observation plugin version, which records the specific version number of the currently integrated tls-observer-dsh SDK, makes it easier to troubleshoot compatibility issues or functional differences between different versions.
service.version	text	The service version number (OpenTelemetry standard Resource field), which identifies the version of the business service that generated this trace or log, such as 1.0.0 or v2.3.1. Used together with service.name, it helps locate which release version an issue belongs to during multi-version deployments.

Torch Log Service