dsh-token-optimizer
Layered token-optimization pipeline for DeepSeek Harness that compresses output, enables MCP lazy loading, and drives proactive compaction.
Install
$ dsh plugin --profile web add ./dsh-token-optimizerPlugin 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
- Converts lengthy text into rendered images for vision-based summarization with dynamic resolution tiers and disk caching
- Applies output ladder formatting to tool outputs (summarizes errors, compresses structured JSON/CSV, samples shell output, and spills large payloads)
- Folds repetitive file reading output into incremental diffs
- Implements tool pruning and lazy-loading of MCP tool schemas to trim per-request token usage
- Proactively triggers core compaction when session token pressure reaches custom thresholds
- Provides aggregated usage monitoring, session status commands, and cache-hit tracking
Useful For
- Optimizing token costs and context window capacity during extended, multi-turn agent sessions
- Managing heavy CLI or tool outputs without overwhelming prompt context limits
- Reducing overhead when integrating multiple MCP servers by lazy-loading tool schemas
Who It Fits
- DeepSeek Harness users and developers managing long-running tasks or token-heavy automated workflows
Documented Limitations
- Native text-to-image rendering relies on Windows PowerShell and .NET System.Drawing; non-Windows environments fallback to disk saving
- Vision summarization requires an accessible DeepSeek API key
- Compaction driver requires compaction services to be active, which may require un-disabling compaction modules on web profile deployments
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, 0 peer, and 0 optional dependencies.
package.json declares DSH bundle metadata.
GitHub reports the repository license as MIT.
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
- Zoria-Lind/dsh-token-optimizer
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin
- Source checked
- 95ca622 · 2026-10-12
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
- 20GitHub stars
- Forks
- 2
- Open issues
- 0
- Last commit
- 2026-10-04
- Last release
- No release detected
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Persistent multi-model agent teams with dynamic DAG orchestration, run monitoring, and token insights for DeepSeek Harness.
View plugin →Cost, token usage, budget, and coding-plan quota tracking plugin for DeepSeek Harness with bilingual UI.
View plugin →Multi-engine search and evidence verification plugin bundle for AI agents in DeepSeek Harness.
View plugin →Generative UI plugin for DeepSeek Harness rendering interactive cards, charts, forms, quizzes, and 3D scenes inline.
View plugin →