64 results
Local-first vision for text-only DeepSeek Harness agents: route image understanding to a local OpenAI-compatible vision model, return structured JSON evidence (summary / OCR / layout / semantics / visual / uncertainty).
Leaderless flat agent teams for DeepSeek Harness: cross-window structured task dispatch, recorder service, and web dashboard
DSH plugin: structured test runner tool (test_run) — auto-detect vitest/jest/pytest/node:test, run tests, parse failure summaries for the model.
Reasonix-style cache-aware compaction backend for DeepSeek Harness (DSH). Replaces/enhances compaction-basic with compact_ratio, one structured summary checkpoint, and a stable recent tail.
One-click health check for a dsh deployment: a sidebar button runs the ops kit's check-health.ps1 (hidden PowerShell) through a plain HTTP route and renders a structured report. The route is independent of the tool registry, so the button still works when agent tool calls break.
DSH standard vision skill: Qwen dynamic-resolution preprocessing + OpenAI-compatible VLM chain with failover/429 backoff, structured evidence mode, local tesseract-first long-screenshot OCR, paste-to-path (no framework patch). 8 tools + runtime skill.
Structured, machine-readable CI output for DeepSeek Harness (dsh): session event capture with JSON/NDJSON reports, JUnit XML, semantic exit codes, artifact collection and privacy redaction. Ships as a dsh profile bundle and a standalone CLI.
dsh-docker — typed, guarded container control for DSH: structured docker/compose tools, project-aware targeting, an approval gate for destructive ops, service-health context, and a replayable status-table renderer
Automated PPT generation for DeepSeek Harness (dsh-plugin): outline + illustrations → editable 16:9 .pptx, HTML preview and structured manifest. 10 built-in visual themes, custom .pptx template upload. Drives DSH's native Univer toolchain.
MindsEye: model-driven vision tools, structured evidence, and exact cache for DeepSeek Harness
Structured code review skill for DeepSeek Harness -- bugs, security, performance, style
Git-backed long-term memory for DeepSeek Harness: a local .mem git repo (single main branch, immutable entries, structured commit messages) with mem_search / mem_read / mem_write / mem_delete / mem_replace tools and fully automatic legacy migration (no CLI needed; the CLI remains for explicit control), auto-managed by the root agent with zero manual steps.
Tencent QQ Agent Mail integration for DeepSeek Harness with a Service, event stream, Web settings, OAuth, and 14 structured tools.
DeepSeek Harness plugin: intercept a page's real data APIs (XHR/fetch JSON) and return structured fields — comments, rankings, prices, tables — with token-efficient field extraction. Requires playwright.
Cloud-backed cross-session memory for DeepSeek Harness: structured entries + vector retrieval persisted to S3-compatible object storage (AWS S3 / MinIO / R2 / OSS), with local cache projection, approval-gated writes and auditable trails
Add selected DSH assistant text to the composer as a structured reference.
Turn-outline sidebar tab for dsh-better-sidebar: fold a session into user turns (input + tool steps + output) with structured jump-back into the conversation | DSH 轮次轨迹侧边栏：按用户轮次折叠会话（输入+工具步骤+输出），一键定位回对话原位
结构化数据提取：从非结构化文本中提取指定字段并输出JSON/CSV/表格
DSH plugin: structured HTTP request tool (http_request) built on the Node built-in fetch.
arXiv search tool plugin for the DeepSeek Harness: query the arXiv preprint API and get structured paper results.
A structured work-state protocol for DeepSeek Harness agent handoffs
Agent governance suite for DeepSeek Harness (enterprise companion): policy-based tool gating (allow/deny/ask), structured JSONL audit trail, and per-agent token quotas against the host token meter — state under $DSH_HOME/gov
Structured context for DeepSeek Harness: split the conversation into typed sub-contexts (reasoning / tool calls / replies) and index them with a table of contents. 结构化 dsh 上下文:按类型拆分小上下文(思考/工具调用/正式回复),以结构化目录为索引。
乔哈里认知四象限对话梳理工具 / Johari cognition-quadrant dialog prompt composer — adds a button above the composer that opens a 2×2 quadrant panel (我知道/不知道 × AI知道/不知道) to groom conversation context into a structured prompt written back into the draft.