← DeepSeek Harness Plugins
DeepSeek Harness PluginManifest Valid

GraphFlow

Token-first surgical code engine and MCP tool provider for AI coding agents with AST parsing, blast-radius guards, and DSH support.

UI & ProductivityDeveloper ToolsBrowser & WebSecurity & PolicySkills & WorkflowsRemote Execution
25GitHub Stars2ForksUpdated2026-10-20

Install

No clear install command was found in the current repository evidence.

Plugin 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.

source-grounded
GraphFlow is a deterministic, token-optimized code knowledge graph and surgical code engine designed for AI coding agents. It exposes four surgical tools (gf_read, gf_edit, gf_exec, gf_macro) backed by true AST parsing for TypeScript/JavaScript via @babel/parser and tree-sitter grammars for Python, Go, Rust, C/C++, and C#. It integrates into DeepSeek Harness via Cordis bundle patches and skill installations.

Key Capabilities

  • Surgical code reading with cross-file type closure, skeleton folding, and receiver-aware blast radius analysis
  • Deterministic in-place and atomic batch edits with cross-repo symbol renaming and structural contract guards
  • Diff-to-test execution routing and dehydrated test output summaries for failure inspection
  • Knowledge graph extraction with defines, imports, calls, and doc-links-code edges
  • DeepSeek Harness integration through Cordis bundle patches and agent skill provisioning

Useful For

  • Navigating codebases and extracting concise symbol slices without blowing context window token budgets
  • Refactoring symbols safely across multiple files with automatic rollback on ambiguous callers
  • Running targeted tests based on git diffs and tsconfig alias configurations with compact error dehydration
  • Supplying DeepSeek Harness coding agents with local knowledge graph queries and habit macros

Who It Fits

  • Developers deploying autonomous coding agents within DeepSeek Harness
  • Software engineers needing token-efficient code context inspection and safe multi-file refactoring
  • Teams seeking local AST-backed code intelligence for agentic workflows

Documented Limitations

  • DSH integration supports MCP tools and skills, but rules and lifecycle hooks are not supported
  • Go same-package imports do not establish file-level imports edges
  • Tree-sitter language grammars are optional native dependencies and revert to heuristic structural parsers when missing
  • Atomic renaming across both Rust traits and their implementations is safely rejected

DSH Compatibility

Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.

Not tested yetNo runtime compatibility tests have been published yet.

Security Signals

Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.

Dependency Counts

package.json declares 3 runtime, 3 development, 0 peer, and 6 optional dependencies.

info
License Declared

GitHub reports the repository license as Apache-2.0.

info
Package Manifest Available

A root package.json was captured and can be inspected by the registry.

info
Source Available

Public GitHub source metadata is available for this registry snapshot.

info

Source & Registry Notes

Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.

Source repository
Roarpeng/GraphFlow
Registry source
GitHub · dsh-plugin topic
Registry classification
Plugin Bundle
Source checked
3cf063b · 2026-10-21

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
25GitHub stars
Forks
2
Open issues
1
Last commit
2026-10-20
Last release
2026-10-20
For maintainers

Maintaining this plugin?

This listing is generated from public repository data. If you maintain this project, you can review the information and share this listing with your users if you find it useful.

Add to README
Listed on DSHPlugin.app