dsh-code-impact
Run the following command in DeepSeek Harness:
dsh plugin install baidd1011/dsh-code-impact
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install baidd1011/dsh-code-impact in DeepSeek Harness to install this plugin; the source repository is available at https://github.com/baidd1011/dsh-code-impact .
About this plugin
In TypeScript and JavaScript projects, tracing a single Git change through the module graph to find affected declarations, files, and tests is slow and error-prone when done by hand. dsh-code-impact performs this analysis in a strictly read-only manner within the DeepSeek Harness: it compares HEAD against the working tree (or any two refs), builds a temporary reverse module dependency graph, and returns changed declarations, impacted files, edge-by-edge propagation paths, file-line evidence, and ranked test candidates. Version 0.2 understands JSONC configs, tsconfig extends, project references, path aliases, and package exports/imports, reusing unchanged snapshots within a single call without maintaining any persistent index, running tests, or writing files.
It suits teams that need a fast impact assessment of TS/JS changes in CI pipelines or local pre-commit hooks, as well as scenarios where a model-reasoning agent needs structured impact context without pulling in a heavyweight static-analysis stack. All results carry explicit truncation markers and warnings; exceeding repository-scale limits returns a clear error code rather than a fabricated partial report.
Use Cases
- Quickly assess TS/JS change impact in CI pipelines
- Provide structured module impact context to reasoning agents
- Locate affected files and ranked test candidates in pre-commit hooks
Best For
- Engineering teams maintaining TS/JS codebases
- Developers building agent workflows in DeepSeek Harness
- Individual developers seeking lightweight impact analysis without persistent indexes
Related Plugins
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.
An armor-breaking plugin for DeepSeek: rewrites instruction priority, expands refusal-blocked phrases, persists across sessions, and shows a green active indicator.