visual-evidence

Here are 5 public repositories matching this topic...

lora-sys / free-vision-skill

A low-token visual evidence compiler for text-only coding agents. Convert images into compact Visual Evidence Packets (VEP) for DeepSeek, Codex, Claude Code, and other text-only models.

ocr opencode vision image-analysis vep codex multimodal deepseek claude-code coding-agent agent-skill visual-evidence low-token

Updated Aug 23, 2026
TypeScript

Argonaut790 / dsh-deepseek-vision

Star

Image understanding, OCR, and persistent visual evidence for text-only DeepSeek Harness models

typescript ocr computer-vision ai-agents image-understanding subagent vision-language-model deepseek multimodal-ai visual-evidence deepseek-harness dsh-plugin

Updated Sep 21, 2026
TypeScript

devdiv07 / when-is-a-screenshot-evidence

Star

Research lab: capture-time provenance and the sufficiency of screenshots as evidence of agent behavior

provenance reproducibility ai-agents agent-evaluation computer-use-agents visual-evidence

Updated Sep 14, 2026
Python

DrAbdulmalek / ocr-core

Star

Shared OCR engine — Visual Evidence Charter (18 rules) + Arabic medical corrections

python ocr medical arabic text-normalization visual-evidence charter-v2

Updated Oct 14, 2026
Python

RNT56 / design-review-workflow

Star

Agentic, evidence-backed website design review workflow for repo-capable AI agents: capture screenshots, inspect visual evidence, and gate business-grade reports.

Updated Aug 1, 2026
TypeScript

Add this topic to your repo

To associate your repository with the visual-evidence topic, visit your repo's landing page and select "manage topics."

Learn more