Let your coding agent SEE what it builds. Cross-platform screenshot feedback for AI agents: an MCP server, a Claude Code hook, and a native DeepSeek Harness plugin that drops screenshots straight into the conversation as images. Python + uv, zero install via uvx. Windows / Linux / macOS.
JARVIS is an intelligent layer that gives agents access to their code, knowledge, context, memory, tools, and runtime — starting locally on your machine, with the ability to extend into the cloud.
Stop AI coding agents from guessing. MCP server + Claude Code plugin that enforces intent alignment BEFORE coding: PRD → intent-confidence gate → Mermaid contracts (state machines / sequence diagrams / decision tables) → mechanical lint with zero CRITICAL → then code. Human-in-the-loop, file-persisted, zero-config.
DeepSeek Harness plugin for local image / voice / music / SFX generation: the same character stays the same character across calls, degenerate output is refused rather than returned, and the GPU is untouched when idle. Hand the image half to any OpenAI-shaped API instead and its engine and 10.1 GiB of weights are never installed.
When I was using agents, I found that most agents rely on curl commands to scrape web pages, which has a very low success rate and is slow, so I made this.
Meta-Forge Workshop: turn recurring work into governed AI workflow assets. MCP server + Claude Code plugin + dsh plugin — staged writes, evals-gated forging, inference engine. 元铸工坊 SOSOFAST：把最佳实践铸成可治理的 AI 工作流资产
Open-source MCP vision server that gives text-only LLMs and AI agents image understanding, OCR, visual analysis, UI inspection, and multimodal capabilities.
Deliverable evidence-chain & conformance checks for DSH — turn requirements into an executable spec, trace every number to a provenance ledger, audit PDF/OOXML metadata and archive manifests. MCP-ready.
brethof-brain client — long-term memory for AI agents: the Claude Code plugin, a Gemini CLI extension and adapters for 20 coding harnesses. Source-available.
Auditable mathematical proving for DeepSeek Harness: a local stdio MCP server (23 tools), a bundled workflow skill, and the /prove and /audit-proof commands. proven / refuted / inconclusive are never conflated, and unproven steps are reported, not hidden.
