Project-Memory-Agent
Project Memory gives coding agents a single, trustworthy memory for a software repository — so they stop re-learning the same facts and stop writing conflicting "memory" files.
2892 results
Project Memory gives coding agents a single, trustworthy memory for a software repository — so they stop re-learning the same facts and stop writing conflicting "memory" files.
给 DeepSeek Harness Web 主窗口加「工作角色」页签 + 对话区像素人办公室浮层(The Agency en / agency-agents-zh 255+ 角色,29 预设团队)。一条命令安装,不改 dsh 源码。
Local-first, durable workflow runtime for Agent Apps—compile static Workflow DSL to LangGraph and orchestrate trusted multi-agent execution through Codex, WorkBuddy, DeepSeek Harness, or any MCP client.
DeepSeek Harness plugin: auto-repairs an invalid messages array (orphaned tool_calls / tool messages) to stop 400 INVALID_REQUEST session lock-ups.
Spritely — a floating work-state mascot for the dsh web UI: it animates with the agent's live activity, tracks the cursor, and offers switchable characters, draggable placement, and customizable backgrounds.
Universal Model Context Protocol (MCP) Bridge Plugin for DeepSeek Harness (dsh)
team-task for DeepSeek Harness: long-horizon multi-agent tasks — a lead plans a reviewed DAG, durable members execute, an event log is the single truth, and a resident reconciler survives crashes and restarts.
An eye that grows: accumulated design taste for DSH agents. A taste corpus + codified principles that sharpen with every example you feed — review designs against learned judgment, not a fixed ruler.
Benchmark-driven self-evolution plugin for DeepSeek Harness: evaluate, optimize, snapshot, accept or roll back.
Agent-driven kanban board plugin for DeepSeek Harness: one board per workspace, per-task agent sessions, human/agent workflow guards, light/dark UI. Based on the dsh-kanban plugin architecture.
Lightweight remote development for DeepSeek Harness: pick a remote workspace over SSH and let the agent work on it with the SAME local tools — file, shell, and search providers are routed to the remote host by the plugin. No extra model tools, no third-party UI plugin.
Server-development mode for DeepSeek Harness: remote agent tools (ssh_list / ssh_exec / ssh_ls / ssh_read / ssh_write) backed by a persistent ssh2 engine, so the DSH coding agent's working environment is a remote Linux server. Works with the `remote` agent preset.
Turn a DeepSeek Harness session into deliverable work reports (daily/weekly/handoff/article) with verifiable receipts
DSH (DeepSeek Harness) workspace archive manager: view archived sessions in place with dimmed titles, filter the workspace list by archived / not-archived from the view options, and unarchive sessions straight from the row menu.
鲸迹:把 DeepSeek Harness Agent 任务生成可分享的竖屏战报视频
Microsoft SkillOpt-Sleep integration for DeepSeek Harness: give your dsh agent a nightly sleep cycle that harvests past sessions, replays recurring tasks, and consolidates validated skills behind a held-out gate.
Enhanced subagent delegation tools for DeepSeek Harness with per-call cwd control: everything in dsh-subagent-tools plus a cwd parameter, shipped with the two in-process provider patches it requires.
Tacit learns what you leave unsaid in your prompts — from messy turns and your own corrections, with zero clicks — and tells the agent how to compensate, on every turn, via a system-prompt section you can read and edit.
A small, session-aware terminal UI for DeepSeek Harness
OpenSpec change orchestration plugin for AI coding agents (Claude Code, Codex, ZCode, DeepSeek Harness, OpenCode)
Send files from your DeepSeek Harness agent straight to your phone, and reference what your phone sent back — no account, no public IP, end-to-end encrypted.
TaskSwarm (蜂群) — multi-agent task orchestration for DeepSeek Harness. Native port of TaskPlane (github.com/HenryLach/taskplane, MIT): waves/lanes parallel execution, git worktree isolation, PROMPT.md/STATUS.md task packets, file mailbox, cross-model review.
Session execution analytics and audit reports for DeepSeek Harness. Understand how your agent actually worked: steps, tool calls, failures, repeated actions, token usage and verification signals.
Observe-only hash-chained evidence receipts for DeepSeek Harness