@yukari316/dsh-toolcall-compat
DSH tool-call compatibility: strips sandbox_permissions/justification from tool-call arguments for third-party models (GPT etc.), and lets the user skip possibly-stuck tool calls with an LLM-facing notice.
1726 results
DSH tool-call compatibility: strips sandbox_permissions/justification from tool-call arguments for third-party models (GPT etc.), and lets the user skip possibly-stuck tool calls with an LLM-facing notice.
Community-maintained floating whale arcade for DeepSeek Harness with five browser-local games
Steam Workshop style plugin store for DSH Web UI: browse, search and one-click install community plugins with mirror acceleration
Session-backed writing pad for the DeepSeek Harness web GUI with Markdown preview, complete-draft XML requests, human-readable transcript projections, and focused AI writing tools.
Eyes for text-only DeepSeek on DeepSeek Harness: image understanding via Antigravity IDE quota (default, flash/pro), OpenAI-compatible VLM endpoints, Gemini API, or local Ollama — model-invokable vision tool, wrapper adapters for deepseek/opencode-go, evidence memory, and a polished client panel.
dsh-buddy — 用户画像自适应表达技能:实时构建用户熟练度/偏好/领域画像,动态匹配回答深度与术语密度
A DeepSeek Harness (DSH) community agent preset for rigorous A-share five-dimension (technical/valuation/fundamental/capital-flow/news) standardized analysis: persona, skill knowledge base, hard risk gate, deterministic Python scoring core, Tencent-first collectors. Not an investment recommendation.
DSH 静态 bundle 插件:定位当前会话中每次通过输入框发送的消息。会话标题栏「消息定位」按钮 → 右侧浮动面板列出全部已发送消息(序号/时间/摘要),支持关键词搜索、点击展开全文、一键回填输入框重新使用。A static-bundle DSH plugin that locates every message you sent from the composer.
Content-aware tool output compression for DeepSeek Harness.
DeepSeek account balance readout in the composer stats line for DSH Web
深海事务所 (Abyss) — an operations console for DeepSeek Harness agent fleets: every agent is a character at a desk, every line comes from the durable session log, and any past case replays and reports from disk
Bounded, read-only design evidence for agent harnesses; compatible with DeepSeek Harness and the Figma REST API.
Give DeepSeek Harness agents the ability to read images directly: a model-facing read_image tool that answers questions about an image through any OpenAI-compatible vision endpoint.
DSH plugin: visible cost/balance stats pill below the composer, with settings page and one-click reset. Fixed fork-level issues of dsh-cost-balance (slot id collision, missing ctx, invisible 5px bar).
An elevated Plugin list for DeepSeek Harness Web Settings — trust tiers, collapsible groups, and complete per-plugin detail
Completion evidence gate for DeepSeek Harness goal-mode subagents
macOS Vision OCR tool plugin for DeepSeek Harness: read text from screen capture, clipboard images, or image files (zh-Hans + en-US). Unofficial community project.
DSH plugin that explicitly forwards the complete Harness environment through the subprocess service
RTK command rewriting suggestions + tool output compaction for DSH. Suggests rtk-rewritten bash commands (deny-mode under rewrite) and compacts noisy tool output (bash/grep/read) to reduce context usage. Port of pi-rtk-optimizer for the Pi coding agent.
Cite what matters in DeepSeek Harness: @ a past session, quote earlier messages with comments, and open the files this turn just wrote.
Volcengine Ark Agent Plan & Coding Plan providers for DeepSeek Harness (DSH), with verified thinking-effort compatibility
A universal hooks compatibility layer: run hooks declared in Claude Code / Codex / opencode / native configs on the DeepSeek Harness, with shell, webhook, oracle and proxy handlers.
Trae-style withdraw for DeepSeek Harness Web: a rollback button on every user/assistant message that restores the workspace code (git commits or hard-link file snapshots) and forks the session back to the last completed turn. 会话撤回插件:每条消息一键回滚代码与对话。
Local vision 'eyes' for DeepSeek Harness (DSH): screen tool (capture screen or image -> local OpenAI-compatible VLM description) and ocr tool (Windows built-in OCR, zero model / GPU / cloud).