DeepSeek Harness plugin exposing MuApi's 100+ image/video/audio generation models as an agent tool
Pending human review — the facts below come from the source repository.
DeepSeek Harness plugin exposing MuApi's 100+ image/video/audio generation models as an agent tool
Users who want this capability in DSH; check the source and docs before installing.
dsh plugin --profile web add github:SamurAIGPT/dsh-plugin-muapi No supported platforms were declared.
muapi_generate tool for MuApi — a unified API for 100+ image, video, audio, and 3D generation models (Flux, Veo 3, Kling, Seedance, Suno, and more) behind a single x-api-key. ## Install sh dsh plugin add @dsh-plugin/dsh-plugin-muapi Then, in Settings → muapi, set your API key credential (create one at muapi.ai/access-keys). ## What it does Registers one tool, muapi_generate: - model — a MuApi model slug, e.g. flux-schnell-image or kling-v2-1-image-to-video. Browse the full catalog at muapi.ai/playground; each model's exact parameters are documented at https://muapi.ai/playground/{model}/llms.txt. - input — that model's parameter object, e.g. { "prompt": "a serene mountain lake at dawn" }. The tool submits the job (POST https://api.muapi.ai/api/v1/{model}), polls GET https://api.muapi.ai/api/v1/predictions/{request_id}/result until it completes or fails, and returns the output URLs. ## Configuration | Setting | Default | Description | | --- | --- | --- | | tool.enabled | true | Expose the tool to the agent. | | tool.apiKeyEnv | — | Credential reference for your MuApi API key. | | tool.baseUrl | https://api.muapi.ai/api/v1 | API base URL. | | tool.pollIntervalMs | 3000 | Delay between result polls. | | tool.timeoutMs | 600000 | Max time to wait for a job to complete. | ## License MITNot by stars — picked by hand, one at a time.
A visual settings dashboard for DeepSeek Harness: every registered settings namespace — including third-party plugin ones the official UI never covers — rendered as editable, schema-driven forms. Early, but the right direction: stop hand-editing YAML.
A DIY trajectory visualizer for DeepSeek Harness: watch the agent’s plan and execution unfold as a graph instead of scrolling raw logs. Early, but it fills a real gap — understanding at a glance what the agent is doing.
Dispatch DSH work from Claude Code / Codex: spin up in-host DSH agent sessions with per-tier presets, watch native subagent progress, and borrow a multimodal bridge that lends the text-only harness vision and image generation. An orchestration plugin that turns DSH into a backend for other coding agents.