dsh-repo-map
Run the following command in DeepSeek Harness:
dsh plugin install MkaliezZ/dsh-repo-map
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install MkaliezZ/dsh-repo-map in DeepSeek Harness to install the plugin; the source code is available at https://github.com/MkaliezZ/dsh-repo-map
About this plugin
Working with AI inside a large codebase, the bottleneck is rarely model intelligence—it is context budget. dsh-repo-map takes a deliberately minimal approach: instead of dumping entire source files into the prompt, it builds a read-only structural and symbol index so the model can survey the project skeleton before diving in. It is designed as a complement to dsh-context-pack; where that plugin handles context injection, dsh-repo-map handles navigation at the structure-and-symbol level, keeping token overlap to a minimum.
Two capabilities do the heavy lifting in v0.1. First, it recognizes common source declarations—functions, classes, variables—and distills them into a symbol map that reflects the shape of your code. Second, it automatically skips generated code and vendor directories, so the index mirrors what you actually wrote rather than what a build system produced. The plugin is lightweight and strictly read-only: it never modifies files and has no write side-effects.
If you frequently need AI assistance in a mid-to-large repository and do not want to burn tokens on third-party dependencies or build artifacts, dsh-repo-map is a low-cost first step: hand the model a map, then let it choose which street to walk down.
Use Cases
- Let AI survey a large repo structure before diving into specifics
- Avoid stuffing entire source files and vendor code into the prompt
- Pair with dsh-context-pack to separate structure awareness from context injection
Best For
- Developers frequently using AI assistance in mid-to-large repositories
- Teams conscious of token budgets and wanting to skip generated code
- DSH users already on dsh-context-pack who need structural navigation
Related Plugins
Memory layer for coding agents that indexes local session history and auto-recalls relevant context before edits, commands, and failures, with no manual search needed.
Traceable, searchable cross-session memory for AI agents that turns conversation knowledge into a typed knowledge graph and recalls relevant subgraphs instead of replaying full history, natively integrated with DeepSeek Harness.
Gives DSH AI cross-session long-term memory, to-do and skill management, plus multi-session orchestration, external AI delegation, and an infinite canvas that grows with you.
dsh-mnemon is a three-tier, pluggable, Agent-driven memory system for DeepSeek Harness, combining Runtime memory, Project Documents, and replaceable Memory Spaces with nine long-term providers.