dsh-vision-router 3.0.3 → 3.0.5

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (35) hide show
  1. package/README.md +183 -162
  2. package/README.zh.md +179 -158
  3. package/cordis.patch.yml +14 -10
  4. package/docs/architecture/dsh-0.2.1-alpha.2-compat.md +41 -0
  5. package/docs/development/dsh-alpha3-compatibility.md +36 -0
  6. package/docs/development/dsh-alpha4-adaptation.md +12 -0
  7. package/docs/development/dsh-alpha4-validation-evidence.md +29 -0
  8. package/docs/development/dsh-alpha5-adaptation.md +31 -0
  9. package/docs/development/dsh-rc1-adaptation.md +29 -0
  10. package/docs/development/dsh-rc2-adaptation.md +24 -0
  11. package/docs/development/v305-real-scenario-acceptance.md +34 -0
  12. package/docs/development/web-live-reload.md +157 -0
  13. package/docs/releases/v3.0.4.md +18 -0
  14. package/docs/releases/v3.0.5.md +31 -0
  15. package/docs/remote-settings.md +4 -0
  16. package/docs/v2-capability-routing.md +25 -9
  17. package/lib/adapter-update-coalescer.js +41 -8
  18. package/lib/client-presentation-boundary-main.js +41 -7
  19. package/lib/client.js +27 -23
  20. package/lib/core-primitives.js +6 -2
  21. package/lib/http-compat.js +33 -2
  22. package/lib/maintenance-routes.js +7 -1
  23. package/lib/session-message-source-compat.js +13 -4
  24. package/lib/sharp-runtime.js +0 -2
  25. package/lib/vision-backend-runtime-policy.js +3 -3
  26. package/lib/vision-background-benchmark.js +33 -5
  27. package/lib/vision-capability-benchmark-service.js +61 -23
  28. package/lib/vision-evidence-guidance.js +7 -1
  29. package/lib/vision-exact-check-client.js +24 -6
  30. package/lib/vision-image-input-verdict.js +5 -3
  31. package/lib/vision-toggle-root-hardening.js +20 -3
  32. package/lib/web-capability-boundary.js +164 -31
  33. package/package.json +6 -4
  34. package/presets/web-connection-ready/index.mjs +48 -0
  35. package/presets/web-connection-ready/package.json +6 -0
package/README.md CHANGED
@@ -9,143 +9,142 @@
9
9
  <p align="center">DeepSeek keeps thinking; the built-in free vision chain and fourteen deep tools do the seeing. When an image matters, enable the composer’s “👁 Vision” control and use image turns like ordinary tool-calling turns — grounded, measurable, repeatable.</p>
10
10
 
11
11
  <p align="center">
12
- <a href="https://awesome-dsh-plugin.com"><img src="https://awesome-dsh-plugin.com/badge.svg" alt="awesome · DSH plugin" /></a>
13
12
  <a href="https://github.com/zp-home/dsh-recommend"><img src="https://img.shields.io/endpoint?url=https%3A%2F%2Fraw.githubusercontent.com%2Fzp-home%2Fdsh-recommend%2Fmain%2Fdata%2Fbadges%2Fysr666__dsh-vision-router.certified.json&amp;style=flat-square" alt="dsh-recommend 🏅 certified" /></a>
14
13
  <a href="https://www.dshbase.com/plugins/dsh-vision-router/"><img src="https://img.shields.io/badge/dshbase-install--tested-2EA44F?style=flat-square" alt="dshbase install-tested" /></a>
15
14
  <a href="https://github.com/SoberReport-AI/DeepGuard/blob/main/reports/dsh-vision-router/2.0.1/39c8f2b2d69aa398418fd6c8ab40b691a92a1a3d.json"><img src="https://img.shields.io/badge/DeepGuard-audit%20clean-2EA44F?style=flat-square" alt="DeepGuard audit clean" /></a>
16
- <a href="https://whyihaveyou.github.io/dsh-suite/"><img src="https://img.shields.io/badge/featured%20on-dsh--suite-4d6bfe?style=flat-square" alt="featured on dsh-suite" /></a>
17
15
  </p>
18
16
 
19
- <p align="center">
20
- <a href="https://github.com/ysr666/dsh-vision-router/releases/tag/v2.3.0"><img src="https://img.shields.io/badge/release-v2.3.0-5B4CF0?style=flat-square" alt="Release v2.3.0" /></a>
21
- <a href="tests"><img src="https://img.shields.io/badge/verified-Node%2022%20%2B%2024-2EA44F?style=flat-square" alt="Verified: Node 22 + 24" /></a>
22
- <a href="LICENSE"><img src="https://img.shields.io/badge/license-MIT-2EA44F?style=flat-square" alt="License: MIT" /></a>
23
- <a href="package.json"><img src="https://img.shields.io/badge/Node.js-%3E%3D22-339933?style=flat-square&amp;logo=nodedotjs&amp;logoColor=white" alt="Node.js >=22" /></a>
24
- <img src="https://img.shields.io/badge/runtime-no%20Python-8A2BE2?style=flat-square" alt="No Python" />
25
- </p>
26
-
27
- <p align="center">
28
- <sub>Ecosystem:</sub>
29
- <a href="https://dshplugin.app/plugins/dsh-vision-router">DSHPlugin.app</a> ·
30
- <a href="https://github.com/diegosouzapw/awesome-omni-dsh-plugins">Awesome Omni DSH Plugins</a> ·
31
- <a href="https://dshpluginhub.ai/plugins/dsh-vision-router">dshpluginhub.ai</a> ·
32
- <a href="https://www.dsh.plus/en/plugins/dsh-vision-router/">dsh.plus</a> ·
33
- <a href="https://dshplugins.ai/">dshplugins.ai</a> ·
34
- <a href="https://dshmarket.com/p/ysr666/dsh-vision-router/">dsh-market</a>
35
- </p>
17
+ <p align="center"><sub>DeepGuard badge links to a historical v2.0.1 audit; it does not certify current releases.</sub></p>
36
18
 
37
19
  <p align="center">English · <a href="README.zh.md">中文</a></p>
38
20
 
39
- <p align="center">💬 <strong>QQ community group: 1105463028</strong></p>
40
-
41
21
  > [!CAUTION]
42
22
  > **Data leaves your machine by default.** When a cloud vision model is used, Vision Router sends the image (or a derived crop), the vision prompt and related request metadata to that provider. A fresh install has an anonymous OVHcloud vision fallback enabled, so "free" and "no key" do **not** mean offline. Do not use the cloud chain for confidential, regulated or classified material. See [Data flow and strict local-only use](#data-flow-and-strict-local-only-use).
43
23
 
44
24
  > [!WARNING]
45
- > 📌 **Announcement (v3.0.3)**
46
- >
47
- > **v3.0.3: presentation and proxy-boundary corrections.** Restored history and empty
48
- > `vision_present` results no longer steal the reader’s scroll position, while fresh presented
49
- > images still focus correctly on both the newer phase-based Host contract and the supported DSH
50
- > 0.1.5 lifecycle. Cookie-domain-style `.example.com` entries now follow the same proxy-host
51
- > policy as `*.example.com`, and the exact DSH 0.1.7 browser smoke again follows both
52
- > presentation-boundary owners. CI action pins were also refreshed; runtime dependencies and
53
- > supported Hosts are unchanged. [What’s new →](docs/releases/v3.0.3.md)
25
+ > **v3.0.5 release notes:** fixes Vision reply continuation, optional-peer activation and OVH request-size handling; improves the Vision toggle's transient catalog-refresh behavior. See [v3.0.5 notes](docs/releases/v3.0.5.md). For the version actually available to install, check [GitHub Releases](https://github.com/ysr666/dsh-vision-router/releases/latest) or npm; release notes are not proof that the version has been published.
26
+
27
+ ## Start here
28
+
29
+ **1. Install in an existing DSH Web profile:**
30
+
31
+ ```sh
32
+ npx @deepseek-ai/dsh plugin --profile web add dsh-vision-router
33
+ ```
34
+
35
+ **2. Keep your usual chat model** and click **👁 Vision** beside the composer; check for **👁 Vision ✓**.
36
+
37
+ **3. Paste or upload an image** and ask the agent what to inspect. The built-in keyless cloud fallback is available out of the box.
38
+
39
+ Using Desktop? Change `--profile web` to `--profile desktop` and reload the app. For Oh-DSH and manual setups, follow the [complete installation guide](#quick-start).
40
+
41
+ [Full quick start](#quick-start) · [What it can do](#highlights) · [Data flow and local-only use](#data-flow-and-strict-local-only-use)
54
42
 
55
43
  <p align="center">
56
44
  <img src="assets/vision-demo.gif" width="640" alt="Demo: paste an image, the agent locates the send button with vision_ground / vision_crop / vision_pixel_diff and answers with coordinates" />
57
45
  </p>
58
46
 
47
+ ## Try it on real images
48
+
49
+ - **Understand a screenshot:** “What does this error dialog mean? Read the exact message.”
50
+ - **Find a UI element:** “Where is the Save button in this screenshot? Give me its position.”
51
+ - **Compare two designs:** “Compare my page to this reference. What should I fix next?” Use `vision_pixel_diff` to measure changes after each iteration.
52
+ - **Read and inspect a long image:** “Extract the relevant text, then crop the unclear part to double-check it.”
53
+
54
+ These are ordinary chat requests: your normal model reasons through the task and calls vision tools as needed. [See the tool list](#tools) or [configure your own/local vision backend](#configuration).
55
+
59
56
  ## Contents
60
57
 
61
- - [Data flow and strict local-only use](#data-flow-and-strict-local-only-use)
62
- - [Why this exists](#why-this-exists)
63
- - [How it compares](#how-it-compares)
64
- - [Design lineage](#design-lineage)
65
- - [Acknowledgements](#acknowledgements)
66
- - [Quick start](#quick-start)
67
- - [Free vision key channels](#free-vision-key-channels)
58
+ - [Try it on real images](#try-it-on-real-images)
68
59
  - [Highlights](#highlights)
69
60
  - [How it works](#how-it-works)
70
61
  - [Tools](#tools)
62
+ - [Quick start](#quick-start)
63
+ - [Data flow and strict local-only use](#data-flow-and-strict-local-only-use)
64
+ - [Free vision key channels](#free-vision-key-channels)
65
+ - [Why this exists](#why-this-exists)
66
+ - [How it compares](#how-it-compares)
71
67
  - [Configuration](#configuration)
72
68
  - [Install and lifecycle](#install-and-lifecycle)
73
69
  - [Troubleshooting](#troubleshooting)
70
+ - [Design lineage](#design-lineage)
71
+ - [Acknowledgements](#acknowledgements)
72
+ - [Release notes and community](#release-notes-and-community)
74
73
 
75
- ## Data flow and strict local-only use
76
-
77
- Vision Router does not operate fully offline in its default configuration. The exact boundary depends on the operation and the selected models:
78
-
79
- | Operation | Where data goes |
80
- |---|---|
81
- | Local pixel tools such as crop, pixel diff, palette, SVG trace, cutout, materialize and HTML screenshot | Processed on the DSH machine; these operations do not call a vision model. |
82
- | `vision_ocr` with Tesseract | Processed on the DSH machine. With the default `auto` engine, an unavailable or empty local OCR result may fall back to a vision model. |
83
- | Image-turn routing and vision-model tools such as describe, detect, ground or vision OCR | The image or derived crop, prompt and relevant context are sent to the selected vision provider. |
84
- | Built-in free fallback | Sent anonymously (no API key) to OVHcloud AI Endpoints at `oai.endpoints.kepler.ai.cloud.ovh.net`; the service still receives the request payload and network metadata such as the source IP. |
85
- | User-configured cloud model / HTTP provider | Sent to that provider's configured endpoint under its own retention and privacy terms. |
86
- | Local Ollama / LM Studio | Image pixels are sent to the configured local endpoint. The resulting text still returns to the current chat model, which may itself be remote. |
74
+ ## Highlights
87
75
 
88
- To stop **Vision Router itself** from sending image data to remote vision endpoints, enable **Settings → Vision Router → General → Local-only vision**. This is a runtime policy, not a destructive rewrite: saved cloud rows and `freeFallback` stay configured but cannot execute while the switch is on. Only loopback visual endpoints (`localhost`, `127.0.0.0/8`, `::1`) are eligible, including Ollama, LM Studio, and custom local HTTP backends. For a **strict local-only workflow**, also use a local chat model because vision result text still returns to the current chat model. Test the final network boundary in your own environment; Vision Router cannot make a remote chat model, proxy or Host integration local.
76
+ - **Capability-aware Auto routing.** Keep configured order for deterministic control, or explicitly enable Auto to prioritize already-configured models using measured capability evidence. Auto never infers capability from model names, and enabling Auto alone does not start benchmarks.
77
+ - **Verifiable model profiling.** Exact Test Vision sends one request to one exact model; Quick and Full benchmark OCR / general / structured / document / grounding capabilities. Background profiling is separately authorized and yields to real foreground vision work.
78
+ - **Original pixels, real answers.** The vision chain reads the image at original resolution (auto-downscaled only to protect latency/quota); the agent's question travels with the image, so answers are about *your* question, not a generic description.
79
+ - **Automatic failover with classified errors.** Region blocks, ToS filtering, 402 quota, 429 rate limits, context overflow, network failures — the chain walks providers one by one and only reports after all of them failed, with actionable advice. A 429 immediately advances to the next backend and opens a Retry-After-aware cooldown instead of sleeping inside the request.
80
+ - **Image memory.** Vision answers are cached by attachment content hash; later text turns substitute the recorded description (marked as untrusted evidence), so DeepSeek genuinely remembers earlier images without re-spending vision calls.
81
+ - **A verifiable pixel loop.** Reference → `vision_html_screenshot` → `vision_pixel_diff` (ratio + worst 8×8-grid regions) → fix → repeat until the mismatch converges. UI restoration becomes measurable instead of eyeballed.
82
+ - **Stable tool schema.** All fourteen deep tools are registered from session start by default, avoiding a mid-conversation tool-list expansion that can invalidate long-context KV/prefix caches. `progressiveTools: true` remains an advanced boot-time opt-in; only then does `vision_activate` mount the tools on demand. See [`docs/progressive-tools-cache.md`](docs/progressive-tools-cache.md).
83
+ - **Selective proxy.** Only the configured vision provider hosts go through your local proxy; DeepSeek stays direct.
89
84
 
90
- ## Why this exists
85
+ ### Pixel loop in practice
91
86
 
92
- Most DSH vision plugins bridge images to DeepSeek as *text descriptions* — lossy, one-shot, and blind to pixels. This plugin keeps the **Host-canonical image pixels on the vision model's side** and DeepSeek on the reasoning side, and makes looking at an image an **ordinary tool call**:
87
+ [![Reference design and final agent rebuild, verified with vision_pixel_diff at 2.54% final difference.](https://raw.githubusercontent.com/ysr666/dsh-vision-router/main/assets/pixel-loop.png)](https://raw.githubusercontent.com/ysr666/dsh-vision-router/main/assets/pixel-loop.png)
93
88
 
94
- - **One command install.** The package ships its own composition patch (`dsh.bundle.patch`): `dsh plugin add` wires the row, the admission wrapper and the attachment limits automatically — zero manual file edits. Taking over the official DeepSeek route is an optional setting (stealth mode, off by default).
95
- - **Free by default.** Vision tools end with a five-model OVHcloud anonymous fallback: no account, no key, 2 requests/minute per IP per model, roughly 10 RPM in theory across independent buckets. User-provided vision models run first.
96
- - **No Python.** The whole pipeline — downscale, grounding, crop, pixel diff, palette, OCR, SVG trace, cutout, HTML screenshot — runs on sharp / potrace / tesseract / system Chrome.
97
- - **Continuous multi-step image work.** An image turn is a text turn that calls tools: `vision_ground` → `vision_crop` → `vision_describe` → `vision_pixel_diff` → fix → screenshot again. The agent keeps iterating until the work is done.
98
- - **DeepSeek stays the brain.** Text turns are untouched in model, cost and context. The vision model is only the eyes, called on demand; answers are cached by image content.
99
- - **Transparent to the user.** Uploaded images keep rendering as images in the conversation UI; the rewrite that points the model at the vision tools happens only inside the model call, never in the session log.
89
+ <p align="center"><sub>Click the image to open the full-resolution original.</sub></p>
100
90
 
101
- ## How it compares
91
+ The agent rebuilt the UI from the reference image, then verified the final result with `vision_pixel_diff`: **2.54% final diff** (32,939 / 1,296,000 differing pixels, threshold 16/channel).
102
92
 
103
- **One-line take**: most dsh vision plugins turn images into *text descriptions* for DeepSeek
104
- (description bridge — lossy); this plugin hands the image turn *straight to a vision model*
105
- (routing bridge — pixel-level), with a built-in keyless free fallback.
93
+ ## How it works
106
94
 
107
- > [!NOTE]
108
- > On DSH 0.1.2-alpha.1+, attachments remain Host-owned. Vision Router consumes the Host-persisted canonical image: clean single-frame 8-bit sRGB/sRGBA images inside the configured normalization limits can pass through byte-identically, while images that need orientation, color-space, metadata, animation, or size normalization may be re-encoded. Pixel tools therefore promise the Host-canonical raster, not preservation of the uploader's original encoded bytes.
95
+ <p align="center">
96
+ <img src="assets/how-it-works.svg" width="100%" alt="How DSH Vision Router keeps DeepSeek as the brain and vision tools as the eyes." />
97
+ </p>
109
98
 
110
- | | Manual model switching | MCP vision bridge | dsh-vision-router |
111
- |---|---|---|---|
112
- | Image pixels | ✅ available (when switched) | ❌ text description only | ✅ Host-canonical raster, on the image turn |
113
- | Automatic | ❌ | ✅ | ✅ |
114
- | Daily model untouched | ❌ (whole session swapped) | ✅ | ✅ |
115
- | Provider failure recovery | ❌ | ❌ | ✅ fallback chains |
116
- | Reusable structured queries | — | partial | ✅ JSON mode + caching |
117
- | Free out-of-the-box | ❌ | ❌ | ✅ built-in keyless endpoint |
118
- | Fits dsh composition | — | external server | ✅ one plugin row |
99
+ The vision model is **only the eyes**; DeepSeek is **always the brain**. An image turn is never hijacked by a one-shot vision answer — the agent drives the tools itself and can keep operating on the image across as many steps as the task needs.
119
100
 
120
- **Difference from existing dsh community projects** (all excellent, each with its own focus; descriptions reflect their READMEs as of 2026-08):
101
+ ## Tools
121
102
 
122
- | Project | Approach | What this plugin adds |
123
- |---|---|---|
124
- | [dsh-vision-sidecar](https://github.com/121103qwq/dsh-vision-sidecar) | Pre-describes images with an external VLM; the description joins the session as a message to DeepSeek; LLM7.io anonymous endpoint by default (OVHcloud listed as a no-key alternative) | Description bridge; this plugin adds image routing, with `vision_describe` covering descriptions on demand |
125
- | [dsh-vision-proxy](https://github.com/Flyvhidbwo/dsh-vision-proxy) | Wraps a provider route and transcribes images into text in the request stream | Transcription bridge; this plugin wraps no provider — it rewrites routing through `agent/request` waterfalls |
126
- | [dsh-vision-provider](https://github.com/libinyam/dsh-vision-provider) | Registers `DeepSeek + Vision` combined routes: images are described by the chosen vision model before reaching DeepSeek | Two-model bridge idea; this plugin adds automatic routing, fallback chains and tools on top |
127
- | [modlens](https://github.com/liustack/modlens) | The first dsh vision plugin; reuses local Claude Code/Codex/OpenCode/Pi logins as vision engines | Engine-reuse idea; this plugin ships its own provider chain and depends on no other local CLI |
128
- | [dsh-vision-toolkit](https://github.com/Anionex/dsh-vision-toolkit) | Ten intent-aware visual tools (Q&A/OCR/pixel verification/UI restoration), called explicitly on demand | Broader tool set; this plugin adds whole-turn auto-routing and a keyless free fallback |
129
- | [dsh-tool-vision](https://github.com/Scorp1o117/dsh-tool-vision) | An `inspect_image` tool plus an `agent/pre-step` waterfall bridge (pasted images become tool hints before entering the log) | Similar waterfall bridge; this plugin adds turn routing, fallback chains, caching and the free endpoint |
103
+ Default `progressiveTools: false`: all fourteen deep tools stay registered from plugin startup, so text and image turns can call them immediately. If you explicitly set `progressiveTools: true` in the profile/composition `cordis.patch.yml`, progressive mode is restored: only `vision_activate` is exposed initially, the full tool set mounts on first use, and the `vision-tools` skill is registered. This is a boot-time switch; restart DSH after changing it. Built on sharp / potrace / tesseract / system Chrome — no Python:
130
104
 
131
- ## Design lineage
105
+ <p align="center">
106
+ <img src="assets/vision-tools.svg" width="100%" alt="Eleven image-processing tools available in DSH Vision Router." />
107
+ </p>
132
108
 
133
- The deep-vision tool layer and UI-restoration workflow in this project were informed by [Anionex/agent-vision-toolkit](https://github.com/Anionex/agent-vision-toolkit) and its native DSH implementation [Anionex/dsh-vision-toolkit](https://github.com/Anionex/dsh-vision-toolkit). In particular, this project drew on their intent-driven tool selection, progressive tool exposure, pixel-diff verification loop, and parts of the visual-tool decomposition and naming, including long-screenshot OCR, foreground extraction, and HTML screenshot tooling.
109
+ The diagram covers the eleven image-processing tools. `vision_present` (durable image delivery) and `vision_bootstrap` (the optional 1+x structured first pass) bring the default deep-tool set to fourteen. Enabling the privacy-gated `vision_screenshot` at boot adds an optional fifteenth tool.
134
110
 
135
- All code in `dsh-vision-router` is independently implemented. On top of those referenced design ideas, this project independently developed its turn-level/tools-first vision routing, DSH admission/wrapper integration, multi-backend provider and failure fallback chains, built-in free vision chain, attachment/image-memory handling, caching, and related runtime resilience mechanisms.
111
+ | Tool | What it does | Artifact |
112
+ |---|---|---|
113
+ | `vision_bootstrap` | Optional 1+x structured first visual pass; establishes task-independent evidence before at least one follow-up vision call | — |
114
+ | `vision_describe` | Image Q&A / multi-image compare / structured-evidence JSON mode (summary + layout regions + entity inventory + verbatim transcription) | — |
115
+ | `vision_materialize` | Copy an authorized attachment into a stable content-addressed workspace handoff file and return absolute + short workspace-relative paths for local OCR/parser fallbacks; no vision/network call | image copy |
116
+ | `vision_ground` | Locate a target → **original-pixel box x1/y1/x2/y2** | annotated PNG (optional) |
117
+ | `vision_detect` | Numbered inventory of every element of a kind (buttons/inputs/links…) with original-pixel boxes | annotated PNG with numbered boxes |
118
+ | `vision_crop` | Crop and zoom into a pixel box | PNG |
119
+ | `vision_present` | Publish a generated or edited local image as a durable chat attachment so the user can see it | image attachment |
120
+ | `vision_pixel_diff` | Per-pixel comparison: diff ratio + worst 8×8-grid regions | red heatmap PNG + JSON report |
121
+ | `vision_colors` | Dominant colors (hex + share) | — |
122
+ | `vision_ocr` | Text transcription: configurable default engine (`auto` / local Tesseract / vision model); an explicit per-call engine still wins | — |
123
+ | `vision_trace` | SVG vectorization (potrace posterization; icons/logos) | SVG |
124
+ | `vision_extract_foreground` | Cutout via border flood fill (uniform backgrounds) | transparent PNG |
125
+ | `vision_html_screenshot` | Screenshot a local HTML file (headless system Chrome); `fullPage: true` captures the whole page and reports `pageHeight` | PNG |
126
+ | `vision_screenshot` | **Disabled by default; explicit opt-in required.** Capture the Windows virtual screen, macOS main display, or Linux root display. Windows uses PowerShell CopyFromScreen, macOS uses `screencapture`, and Linux requires ImageMagick `import` or `scrot`; optional `identify=true` tries enabled local recognition backends in order and returns the description with the path | PNG / + description |
127
+ | `vision_long_screenshot_ocr` | Long-screenshot transcription: overlapping chunks, tesseract first / vision model fallback, stitched Markdown | chunk PNGs + Markdown + manifest |
136
128
 
137
- We appreciate Anionex's prior work and the broader DSH community. Clear attribution and independent iteration can coexist; both help keep the DSH ecosystem open, collaborative, and healthy.
129
+ Formats are sniffed from magic bytes, so extensionless content-addressed attachment files work everywhere (no `.png` renaming needed).
138
130
 
139
- ## Acknowledgements
131
+ **Common workflows**
140
132
 
141
- This project borrows ideas from all of the above — especially the keyless free-endpoint
142
- exploration (LLM7.io and OVHcloud anonymous tiers) by
143
- [dsh-vision-sidecar](https://github.com/121103qwq/dsh-vision-sidecar). Thanks to the authors of
144
- [dsh-vision-proxy](https://github.com/Flyvhidbwo/dsh-vision-proxy),
145
- [dsh-vision-provider](https://github.com/libinyam/dsh-vision-provider),
146
- [modlens](https://github.com/liustack/modlens),
147
- [dsh-vision-toolkit](https://github.com/Anionex/dsh-vision-toolkit), and
148
- [dsh-tool-vision](https://github.com/Scorp1o117/dsh-tool-vision).
133
+ ```text
134
+ vision_ground image="ref.png" target="the send button"
135
+ vision_detect image="page.png" target="input fields"
136
+ vision_crop image="ref.png" region="1067,841,1108,881"
137
+ vision_present path="rebuilt.png"
138
+ vision_describe paths=["ref.png","impl.png"] question="list the differences" json=true
139
+ vision_pixel_diff original="ref.png" rebuilt="screenshot.png"
140
+ vision_ocr image="screenshot.png"
141
+ vision_colors image="ref.png" top=8
142
+ vision_trace image="icon.png" steps=4
143
+ vision_extract_foreground image="logo.png"
144
+ vision_html_screenshot source="page.html" width=1200 height=720
145
+ vision_html_screenshot source="page.html" width=1200 height=720 fullPage=true
146
+ vision_long_screenshot_ocr image="chat-log.png" chunkHeight=1200 overlap=120
147
+ ```
149
148
 
150
149
  ## Quick start
151
150
 
@@ -207,7 +206,7 @@ With “👁 Vision” enabled, paste or upload an image normally. By default th
207
206
 
208
207
  If the session already contains images, DSH may reject switching from a vision wrapper back to a text-only route. Vision Router does not bypass that Host rule: it shows a transient error using the same interaction style as the stock model selector, keeps the real current model unchanged, and leaves `👁 Vision ✓` reflecting the actual state so the session stays usable.
209
208
 
210
- The built-in anonymous OVH vision fallback is already configured, so normal image use needs no signup or API key. **The lower-right chat picker selects only the brain/conversation model**; vision backends do not belong there. Advanced options live under **Settings → Vision Router**: each vision-backend row may select any callable generative user model already configured under **Settings → Models**. DSH image-capability metadata is advisory only: undeclared or text-only-labelled models remain selectable and show a warning. At runtime Vision Router always tries the provider's registered DSH adapter first — including WebSocket, RPC and private transports — and falls through on a real failure. The direct compatibility bridge is used only when an http(s) OpenAI Chat Completions endpoint is positively identified. Leaving every user row empty is valid; the OVH chain remains the final fallback. `Vision HTTP` is an internal transport route, not a model group users should select.
209
+ The built-in anonymous OVH vision fallback is already configured, so normal image use needs no signup or API key. **The lower-right chat picker selects only the brain/conversation model**; vision backends do not belong there. Advanced options live under **Settings → Vision Router**: each vision-backend row may select any callable generative user model already configured under **Settings → Models**. DSH image-capability metadata is advisory for backend selection, but an explicit text-only declaration makes the Host replace image pixels with text placeholders before adapter dispatch. Vision Router uses a direct pixel bridge only for verified http(s) OpenAI Chat Completions endpoints; otherwise that backend fails safely and the chain continues. WebSocket, RPC, Responses API and other native protocols use their registered adapters when image input is declared or modalities are unknown. If a custom model genuinely supports images, first enable **Image** under **Settings → Models → provider editor → Customized settings → Model options → Input types** when that Host control is available. On older Hosts without the control, set the provider-specific model field (`llm-pi-ai`: `input: [text, image]`; `llm-deepseek`: `inputModalities: [text, image]`) and reload the model configuration. Never claim image input for an endpoint that cannot accept images. Leaving every user row empty is valid; the OVH chain remains the final fallback. `Vision HTTP` is an internal transport route, not a model group users should select.
211
210
 
212
211
  ### See it in action
213
212
 
@@ -218,6 +217,21 @@ The built-in anonymous OVH vision fallback is already configured, so normal imag
218
217
  <img src="assets/dsh-conversation-image-qa-result.png" width="49%" alt="The agent's structured answer describing the image content." />
219
218
  </p>
220
219
 
220
+ ## Data flow and strict local-only use
221
+
222
+ Vision Router does not operate fully offline in its default configuration. The exact boundary depends on the operation and the selected models:
223
+
224
+ | Operation | Where data goes |
225
+ |---|---|
226
+ | Local pixel tools such as crop, pixel diff, palette, SVG trace, cutout, materialize and HTML screenshot | Processed on the DSH machine; these operations do not call a vision model. |
227
+ | `vision_ocr` with Tesseract | Processed on the DSH machine. With the default `auto` engine, an unavailable or empty local OCR result may fall back to a vision model. |
228
+ | Image-turn routing and vision-model tools such as describe, detect, ground or vision OCR | The image or derived crop, prompt and relevant context are sent to the selected vision provider. |
229
+ | Built-in free fallback | Sent anonymously (no API key) to OVHcloud AI Endpoints at `oai.endpoints.kepler.ai.cloud.ovh.net`; the service still receives the request payload and network metadata such as the source IP. |
230
+ | User-configured cloud model / HTTP provider | Sent to that provider's configured endpoint under its own retention and privacy terms. |
231
+ | Local Ollama / LM Studio | Image pixels are sent to the configured local endpoint. The resulting text still returns to the current chat model, which may itself be remote. |
232
+
233
+ To stop **Vision Router itself** from sending image data to remote vision endpoints, enable **Settings → Vision Router → General → Local-only vision**. This is a runtime policy, not a destructive rewrite: saved cloud rows and `freeFallback` stay configured but cannot execute while the switch is on. Only loopback visual endpoints (`localhost`, `127.0.0.0/8`, `::1`) are eligible, including Ollama, LM Studio, and custom local HTTP backends. For a **strict local-only workflow**, also use a local chat model because vision result text still returns to the current chat model. Test the final network boundary in your own environment; Vision Router cannot make a remote chat model, proxy or Host integration local.
234
+
221
235
  ## Free vision key channels
222
236
 
223
237
  The built-in OVH fallback is anonymous by design, and OVH caps anonymous use at **2 requests/minute per IP per model**. If that feels tight, every channel below offers **free vision models with much higher quotas** — all of them are free to register, and none charges for the free tier. Free policies rotate often; treat this table as an August 2026 snapshot and double-check each provider's console before relying on it.
@@ -239,80 +253,46 @@ Any of these channels can join the vision chain as an `httpProviders` entry (key
239
253
  > [!NOTE]
240
254
  > Free-tier policies change without notice — Cerebras retired its free tier in July 2026 (now a one-time $5 credit), SambaNova's free tier is down to 20 requests/day, and Hugging Face's is $0.10/month. Third-party “`:free` relay” aggregators are deliberately not listed: they rotate quickly, lack SLAs, and some resell quota in ways that violate upstream terms.
241
255
 
242
- ## Highlights
243
-
244
- - **Capability-aware Auto routing.** Keep configured order for deterministic control, or explicitly enable Auto to prioritize already-configured models using measured capability evidence. Auto never infers capability from model names, and enabling Auto alone does not start benchmarks.
245
- - **Verifiable model profiling.** Exact Test Vision sends one request to one exact model; Quick and Full benchmark OCR / general / structured / document / grounding capabilities. Background profiling is separately authorized and yields to real foreground vision work.
246
- - **Original pixels, real answers.** The vision chain reads the image at original resolution (auto-downscaled only to protect latency/quota); the agent's question travels with the image, so answers are about *your* question, not a generic description.
247
- - **Automatic failover with classified errors.** Region blocks, ToS filtering, 402 quota, 429 rate limits, context overflow, network failures — the chain walks providers one by one and only reports after all of them failed, with actionable advice. A 429 immediately advances to the next backend and opens a Retry-After-aware cooldown instead of sleeping inside the request.
248
- - **Image memory.** Vision answers are cached by attachment content hash; later text turns substitute the recorded description (marked as untrusted evidence), so DeepSeek genuinely remembers earlier images without re-spending vision calls.
249
- - **A verifiable pixel loop.** Reference → `vision_html_screenshot` → `vision_pixel_diff` (ratio + worst 8×8-grid regions) → fix → repeat until the mismatch converges. UI restoration becomes measurable instead of eyeballed.
250
- - **Stable tool schema.** All fourteen deep tools are registered from session start by default, avoiding a mid-conversation tool-list expansion that can invalidate long-context KV/prefix caches. `progressiveTools: true` remains an advanced boot-time opt-in; only then does `vision_activate` mount the tools on demand. See [`docs/progressive-tools-cache.md`](docs/progressive-tools-cache.md).
251
- - **Selective proxy.** Only the configured vision provider hosts go through your local proxy; DeepSeek stays direct.
252
-
253
- ### Pixel loop in practice
254
-
255
- [![Reference design and final agent rebuild, verified with vision_pixel_diff at 2.54% final difference.](https://raw.githubusercontent.com/ysr666/dsh-vision-router/main/assets/pixel-loop.png)](https://raw.githubusercontent.com/ysr666/dsh-vision-router/main/assets/pixel-loop.png)
256
-
257
- <p align="center"><sub>Click the image to open the full-resolution original.</sub></p>
258
-
259
- The agent rebuilt the UI from the reference image, then verified the final result with `vision_pixel_diff`: **2.54% final diff** (32,939 / 1,296,000 differing pixels, threshold 16/channel).
256
+ ## Why this exists
260
257
 
261
- ## How it works
258
+ Most DSH vision plugins bridge images to DeepSeek as *text descriptions* — lossy, one-shot, and blind to pixels. This plugin keeps the **Host-canonical image pixels on the vision model's side** and DeepSeek on the reasoning side, and makes looking at an image an **ordinary tool call**:
262
259
 
263
- <p align="center">
264
- <img src="assets/how-it-works.svg" width="100%" alt="How DSH Vision Router keeps DeepSeek as the brain and vision tools as the eyes." />
265
- </p>
260
+ - **One command install.** The package ships its own composition patch (`dsh.bundle.patch`): `dsh plugin add` wires the row, the admission wrapper and the attachment limits automatically — zero manual file edits. Taking over the official DeepSeek route is an optional setting (stealth mode, off by default).
261
+ - **Free by default.** Vision tools end with a four-model OVHcloud anonymous fallback: no account, no key, 2 requests/minute per IP per model, roughly 8 RPM in theory across independent buckets. User-provided vision models run first.
262
+ - **No Python.** The whole pipeline — downscale, grounding, crop, pixel diff, palette, OCR, SVG trace, cutout, HTML screenshot — runs on sharp / potrace / tesseract / system Chrome.
263
+ - **Continuous multi-step image work.** An image turn is a text turn that calls tools: `vision_ground` → `vision_crop` → `vision_describe` → `vision_pixel_diff` → fix → screenshot again. The agent keeps iterating until the work is done.
264
+ - **DeepSeek stays the brain.** Text turns are untouched in model, cost and context. The vision model is only the eyes, called on demand; answers are cached by image content.
265
+ - **Transparent to the user.** Uploaded images keep rendering as images in the conversation UI; the rewrite that points the model at the vision tools happens only inside the model call, never in the session log.
266
266
 
267
- The vision model is **only the eyes**; DeepSeek is **always the brain**. An image turn is never hijacked by a one-shot vision answer — the agent drives the tools itself and can keep operating on the image across as many steps as the task needs.
267
+ ## How it compares
268
268
 
269
- ## Tools
269
+ **One-line take**: most dsh vision plugins turn images into *text descriptions* for DeepSeek
270
+ (description bridge — lossy); this plugin hands the image turn *straight to a vision model*
271
+ (routing bridge — pixel-level), with a built-in keyless free fallback.
270
272
 
271
- Default `progressiveTools: false`: all fourteen deep tools stay registered from plugin startup, so text and image turns can call them immediately. If you explicitly set `progressiveTools: true` in the profile/composition `cordis.patch.yml`, progressive mode is restored: only `vision_activate` is exposed initially, the full tool set mounts on first use, and the `vision-tools` skill is registered. This is a boot-time switch; restart DSH after changing it. Built on sharp / potrace / tesseract / system Chrome — no Python:
273
+ > [!NOTE]
274
+ > On DSH 0.1.2-alpha.1+, attachments remain Host-owned. Vision Router consumes the Host-persisted canonical image: clean single-frame 8-bit sRGB/sRGBA images inside the configured normalization limits can pass through byte-identically, while images that need orientation, color-space, metadata, animation, or size normalization may be re-encoded. Pixel tools therefore promise the Host-canonical raster, not preservation of the uploader's original encoded bytes.
272
275
 
273
- <p align="center">
274
- <img src="assets/vision-tools.svg" width="100%" alt="Eleven image-processing tools available in DSH Vision Router." />
275
- </p>
276
+ | | Manual model switching | MCP vision bridge | dsh-vision-router |
277
+ |---|---|---|---|
278
+ | Image pixels | ✅ available (when switched) | ❌ text description only | ✅ Host-canonical raster, on the image turn |
279
+ | Automatic | ❌ | ✅ | ✅ |
280
+ | Daily model untouched | ❌ (whole session swapped) | ✅ | ✅ |
281
+ | Provider failure recovery | ❌ | ❌ | ✅ fallback chains |
282
+ | Reusable structured queries | — | partial | ✅ JSON mode + caching |
283
+ | Free out-of-the-box | ❌ | ❌ | ✅ built-in keyless endpoint |
284
+ | Fits dsh composition | — | external server | ✅ one plugin row |
276
285
 
277
- The diagram covers the eleven image-processing tools. `vision_present` (durable image delivery) and `vision_bootstrap` (the optional 1+x structured first pass) bring the default deep-tool set to fourteen. Enabling the privacy-gated `vision_screenshot` at boot adds an optional fifteenth tool.
286
+ **Difference from existing dsh community projects** (all excellent, each with its own focus; descriptions reflect their READMEs as of 2026-08):
278
287
 
279
- | Tool | What it does | Artifact |
288
+ | Project | Approach | What this plugin adds |
280
289
  |---|---|---|
281
- | `vision_bootstrap` | Optional 1+x structured first visual pass; establishes task-independent evidence before at least one follow-up vision call | — |
282
- | `vision_describe` | Image Q&A / multi-image compare / structured-evidence JSON mode (summary + layout regions + entity inventory + verbatim transcription) | — |
283
- | `vision_materialize` | Copy an authorized attachment into a stable content-addressed workspace handoff file and return absolute + short workspace-relative paths for local OCR/parser fallbacks; no vision/network call | image copy |
284
- | `vision_ground` | Locate a target → **original-pixel box x1/y1/x2/y2** | annotated PNG (optional) |
285
- | `vision_detect` | Numbered inventory of every element of a kind (buttons/inputs/links…) with original-pixel boxes | annotated PNG with numbered boxes |
286
- | `vision_crop` | Crop and zoom into a pixel box | PNG |
287
- | `vision_present` | Publish a generated or edited local image as a durable chat attachment so the user can see it | image attachment |
288
- | `vision_pixel_diff` | Per-pixel comparison: diff ratio + worst 8×8-grid regions | red heatmap PNG + JSON report |
289
- | `vision_colors` | Dominant colors (hex + share) | — |
290
- | `vision_ocr` | Text transcription: configurable default engine (`auto` / local Tesseract / vision model); an explicit per-call engine still wins | — |
291
- | `vision_trace` | SVG vectorization (potrace posterization; icons/logos) | SVG |
292
- | `vision_extract_foreground` | Cutout via border flood fill (uniform backgrounds) | transparent PNG |
293
- | `vision_html_screenshot` | Screenshot a local HTML file (headless system Chrome); `fullPage: true` captures the whole page and reports `pageHeight` | PNG |
294
- | `vision_screenshot` | **Disabled by default; explicit opt-in required.** Capture the Windows virtual screen, macOS main display, or Linux root display. Windows uses PowerShell CopyFromScreen, macOS uses `screencapture`, and Linux requires ImageMagick `import` or `scrot`; optional `identify=true` tries enabled local recognition backends in order and returns the description with the path | PNG / + description |
295
- | `vision_long_screenshot_ocr` | Long-screenshot transcription: overlapping chunks, tesseract first / vision model fallback, stitched Markdown | chunk PNGs + Markdown + manifest |
296
-
297
- Formats are sniffed from magic bytes, so extensionless content-addressed attachment files work everywhere (no `.png` renaming needed).
298
-
299
- **Common workflows**
300
-
301
- ```text
302
- vision_ground image="ref.png" target="the send button"
303
- vision_detect image="page.png" target="input fields"
304
- vision_crop image="ref.png" region="1067,841,1108,881"
305
- vision_present path="rebuilt.png"
306
- vision_describe paths=["ref.png","impl.png"] question="list the differences" json=true
307
- vision_pixel_diff original="ref.png" rebuilt="screenshot.png"
308
- vision_ocr image="screenshot.png"
309
- vision_colors image="ref.png" top=8
310
- vision_trace image="icon.png" steps=4
311
- vision_extract_foreground image="logo.png"
312
- vision_html_screenshot source="page.html" width=1200 height=720
313
- vision_html_screenshot source="page.html" width=1200 height=720 fullPage=true
314
- vision_long_screenshot_ocr image="chat-log.png" chunkHeight=1200 overlap=120
315
- ```
290
+ | [dsh-vision-sidecar](https://github.com/121103qwq/dsh-vision-sidecar) | Pre-describes images with an external VLM; the description joins the session as a message to DeepSeek; LLM7.io anonymous endpoint by default (OVHcloud listed as a no-key alternative) | Description bridge; this plugin adds image routing, with `vision_describe` covering descriptions on demand |
291
+ | [dsh-vision-proxy](https://github.com/Flyvhidbwo/dsh-vision-proxy) | Wraps a provider route and transcribes images into text in the request stream | Transcription bridge; this plugin wraps no provider — it rewrites routing through `agent/request` waterfalls |
292
+ | [dsh-vision-provider](https://github.com/libinyam/dsh-vision-provider) | Registers `DeepSeek + Vision` combined routes: images are described by the chosen vision model before reaching DeepSeek | Two-model bridge idea; this plugin adds automatic routing, fallback chains and tools on top |
293
+ | [modlens](https://github.com/liustack/modlens) | The first dsh vision plugin; reuses local Claude Code/Codex/OpenCode/Pi logins as vision engines | Engine-reuse idea; this plugin ships its own provider chain and depends on no other local CLI |
294
+ | [dsh-vision-toolkit](https://github.com/Anionex/dsh-vision-toolkit) | Ten intent-aware visual tools (Q&A/OCR/pixel verification/UI restoration), called explicitly on demand | Broader tool set; this plugin adds whole-turn auto-routing and a keyless free fallback |
295
+ | [dsh-tool-vision](https://github.com/Scorp1o117/dsh-tool-vision) | An `inspect_image` tool plus an `agent/pre-step` waterfall bridge (pasted images become tool hints before entering the log) | Similar waterfall bridge; this plugin adds turn routing, fallback chains, caching and the free endpoint |
316
296
 
317
297
  ## Provider fallback chain
318
298
 
@@ -322,7 +302,7 @@ The vision tools try backends in order and surface an error only after all of th
322
302
  2. **Local Ollama (optional, off by default)**: `localOllama.enabled` adds keyless, offline recognition through your local Ollama (for example qwen2.5vl);
323
303
  3. **Local LM Studio (optional, off by default)**: `localLmStudio.enabled` follows Ollama and requires the real model identifier shown in LM Studio Developer or returned by `/v1/models`;
324
304
  4. **Advanced custom HTTP vision endpoints**: legacy/advanced `httpProviders`, when present, run after the local backends;
325
- 5. **Built-in anonymous OVH fallback**: always last and never exposed in a model picker. The current quality-first chain is `Qwen3.5-397B-A17B` → `Qwen2.5-VL-72B-Instruct` → `Qwen3.6-27B` → `Mistral-Small-3.2-24B-Instruct-2506` → `Qwen3.5-9B`. OVH anonymous limits are **2 requests/minute per IP per model**. The five models have independent buckets, so spreading requests across them is about **10 RPM in theory**, subject to OVH's actual rate limiting. No signup or API key is required. Want more headroom? See [Free vision key channels](#free-vision-key-channels) — a free OVH access key lifts this same endpoint to 400 requests/minute.
305
+ 5. **Built-in anonymous OVH fallback**: always last and never exposed in a model picker. The current quality-first chain is `Qwen3.5-397B-A17B` → `Qwen2.5-VL-72B-Instruct` → `Qwen3.6-27B` → `Qwen3.5-9B`. OVH anonymous limits are **2 requests/minute per IP per model**. The four models have independent buckets, so spreading requests across them is about **8 RPM in theory**, subject to OVH's actual rate limiting. No signup or API key is required. OVH has scheduled the previously bundled Mistral Small 3.2 model for decommission on November 4, 2026; it is no longer part of the automatic fallback, but user-managed rows are not changed. Want more headroom? See [Free vision key channels](#free-vision-key-channels) — a free OVH access key lifts this same endpoint to 400 requests/minute.
326
306
 
327
307
  > [!IMPORTANT]
328
308
  > This “vision chain” is the **eyes** used by Vision Router: each settings row selects one user vision model, while the lower-right chat picker selects the **brain/conversation model**. The two are deliberately separate. Text-only DeepSeek/opencode models are filtered out of the vision-backend dropdown, and the internal `Vision HTTP` transport route is no longer exposed to users.
@@ -356,6 +336,8 @@ The settings card uses provider + model dropdowns; an empty model means every mo
356
336
 
357
337
  The Web profile registers a first-class **Settings → Vision Router** surface. Its General page keeps model choice and v2 routing authority together; Vision Strategy, Local & Device, Advanced and Diagnostics separate tool behavior, local backends, sensitive/performance controls and troubleshooting.
358
338
 
339
+ **Reverse proxy deployments:** Preserve the original browser-facing `Host` when forwarding to DSH, as required by [upstream DSH](https://github.com/deepseek-ai/deepseek-harness/blob/d743267388641bc76f17c45ce8b4c231aed1d32c/docs/user/guide/public-deployments.md). A remote request forwarded over loopback must not become a *local-machine-only* request. Proxy-marked loopback requests are denied local-only Vision Router capabilities; proxies that rewrite `Host` to localhost **and strip all forwarding metadata cannot be reliably identified by the current Host API**. See [remote settings security](docs/remote-settings.md) and [Issue #680](https://github.com/ysr666/dsh-vision-router/issues/680).
340
+
359
341
  - **Vision model chain**: the real image-capable models used by `vision_describe` and friends; the built-in free chain remains the final fallback;
360
342
  - **Model selection**: keep the configured order, or explicitly enable capability-aware Auto with Balanced / Quality / Speed / Local preference;
361
343
  - **Background capability data**: `off`, `local-free`, or `all`; this is separately authorized and never turns on merely because Auto was enabled;
@@ -460,6 +442,8 @@ ollama pull qwen2.5vl
460
442
 
461
443
  ## Install and lifecycle
462
444
 
445
+ **Source developers:** use [Web development watch, Live Reload and native HMR](docs/development/web-live-reload.md) with an isolated DSH profile. This workflow is separate from packaged installation and does not enable Desktop HMR.
446
+
463
447
  ### Install
464
448
 
465
449
  Normal npm/npx install — one command:
@@ -617,6 +601,43 @@ Manual fallback: in VS Code, use “Save with Encoding” → `UTF-8` (without B
617
601
  - Secrets never travel: `apiKeyEnv` names a DSH credential reference; the value is resolved per call and never logged.
618
602
  - The settings write path goes through the settings service (schema-validated, revision-checked) — a stale or invalid save is rejected, not partially applied.
619
603
 
604
+ ## Design lineage
605
+
606
+ The deep-vision tool layer and UI-restoration workflow in this project were informed by [Anionex/agent-vision-toolkit](https://github.com/Anionex/agent-vision-toolkit) and its native DSH implementation [Anionex/dsh-vision-toolkit](https://github.com/Anionex/dsh-vision-toolkit). In particular, this project drew on their intent-driven tool selection, progressive tool exposure, pixel-diff verification loop, and parts of the visual-tool decomposition and naming, including long-screenshot OCR, foreground extraction, and HTML screenshot tooling.
607
+
608
+ All code in `dsh-vision-router` is independently implemented. On top of those referenced design ideas, this project independently developed its turn-level/tools-first vision routing, DSH admission/wrapper integration, multi-backend provider and failure fallback chains, built-in free vision chain, attachment/image-memory handling, caching, and related runtime resilience mechanisms.
609
+
610
+ We appreciate Anionex's prior work and the broader DSH community. Clear attribution and independent iteration can coexist; both help keep the DSH ecosystem open, collaborative, and healthy.
611
+
612
+ ## Acknowledgements
613
+
614
+ This project borrows ideas from all of the above — especially the keyless free-endpoint
615
+ exploration (LLM7.io and OVHcloud anonymous tiers) by
616
+ [dsh-vision-sidecar](https://github.com/121103qwq/dsh-vision-sidecar). Thanks to the authors of
617
+ [dsh-vision-proxy](https://github.com/Flyvhidbwo/dsh-vision-proxy),
618
+ [dsh-vision-provider](https://github.com/libinyam/dsh-vision-provider),
619
+ [modlens](https://github.com/liustack/modlens),
620
+ [dsh-vision-toolkit](https://github.com/Anionex/dsh-vision-toolkit), and
621
+ [dsh-tool-vision](https://github.com/Scorp1o117/dsh-tool-vision).
622
+
623
+ ## Release notes and community
624
+
625
+ Release details: [v3.0.5 release notes](docs/releases/v3.0.5.md). Check GitHub Releases or npm for the published version.
626
+
627
+ <p align="center">
628
+ <sub>Ecosystem:</sub>
629
+ <a href="https://dshplugin.app/plugins/dsh-vision-router">DSHPlugin.app</a> ·
630
+ <a href="https://github.com/diegosouzapw/awesome-omni-dsh-plugins">Awesome Omni DSH Plugins</a> ·
631
+ <a href="https://dshpluginhub.ai/plugins/dsh-vision-router">dshpluginhub.ai</a> ·
632
+ <a href="https://www.dsh.plus/en/plugins/dsh-vision-router/">dsh.plus</a> ·
633
+ <a href="https://dshplugins.ai/">dshplugins.ai</a> ·
634
+ <a href="https://dshmarket.com/p/ysr666/dsh-vision-router/">dsh-market</a>
635
+ </p>
636
+
637
+ <p align="center">Repository: <a href="https://github.com/ysr666/dsh-vision-router">GitHub (primary · issues, PRs and releases)</a> · <a href="https://atomgit.com/ysr666/dsh-vision-router">AtomGit (China mirror)</a></p>
638
+
639
+ <p align="center">💬 <strong>QQ community group: 1105463028</strong></p>
640
+
620
641
  ## License
621
642
 
622
643
  [MIT](LICENSE)