Install
$ dsh plugin --profile web add dsh-llm-tracePlugin Overview & Capabilities
AI-assisted organization based on the public repository snapshot. The content must be grounded in source evidence and does not replace compatibility or security verification.
Key Capabilities
- Captures exact wire-level HTTP request and response bodies by wrapping globalThis.fetch without blocking streaming responses
- Integrates a session-scoped LLM Trace tab into the Conversation View UI with automated composer toggling
- Provides a global loopback web viewer endpoint for inspecting all process-wide model calls
- Renders massive JSON payloads as collapsible trees with measured byte sizes and text expansion
- Assembles streaming Server-Sent Events (SSE) deltas back into aggregated reasoning, text content, and tool calls
- Configurable header redaction, URL pattern matching, retention counts, and journal memory limits
Useful For
- Debugging exact provider payloads, prompt formatting, and token bloat during agent interactions
- Inspecting background subagent or title-generator requests across the entire Harness process
- Analyzing detailed reasoning output, raw tool calling parameters, and streaming chunk structures
- Measuring exact byte distribution across message arrays and provider envelopes
Who It Fits
- Developers debugging DeepSeek Harness agent workflows and model interactions
- LLM application engineers diagnosing prompt payloads, tool schemas, and streaming latency
- Researchers monitoring low-level model provider network exchanges
Documented Limitations
- Does not redact sensitive credentials or secrets within request bodies (only headers are redacted)
- Non-default mount paths require manual matching adjustments in the client script
- Relies on wrapping global fetch; conflicts or duplicates traces if run alongside other global fetch wrappers
- SSE response reassembly currently targets OpenAI-formatted streaming envelopes, falling back to raw view for others
DSH Compatibility
Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.
Security Signals
Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.
package.json declares 0 runtime, 0 development, 1 peer, and 0 optional dependencies.
package.json declares DSH bundle metadata.
GitHub reports the repository license as Apache-2.0.
A root package.json was captured and can be inspected by the registry.
Public GitHub source metadata is available for this registry snapshot.
Source & Registry Notes
Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.
- Source repository
- wpeng77/dsh-llm-trace
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin
- Source checked
- 9fffd23 · 2026-09-24
This project is independently indexed from public source information. DSH Plugin is not affiliated with DeepSeek or the plugin author. Always check the author repository before installation.
Repository Activity
- GitHub stars
- 0GitHub stars
- Forks
- 0
- Open issues
- 0
- Last commit
- 2026-09-21
- Last release
- No release detected
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Session lifecycle manager for DeepSeek Harness with search, restore, idle cleanup, and diagnosis.
View plugin →Persistent multi-model agent teams with dynamic DAG orchestration, run monitoring, and token insights for DeepSeek Harness.
View plugin →Agent skill for DeepSeek Harness that turns ideas, plans, and codebases into interactive architecture diagrams.
View plugin →Native SQLite-backed task management and project board bundle with Web UI and Agent workflow tools for DeepSeek Harness.
View plugin →