32 results
Cross-session project memory for DeepSeek Harness: seven-layer SQLite memory, first-turn snapshot injection, per-message keyword hits, memory_remember/search/project tools, automatic reflection with reflection-fold UI, and idle-triggered dream consolidati
DSH web extension: a live subagent run monitor with a sidebar trigger and a floating card panel
DeepSeek Harness UX walkthrough plugin: persona-driven source-code UX review for React (TypeScript/JavaScript) and Vue 3 projects, with change-triggered automatic walkthroughs
Self-contained Trellis workflow trigger for DeepSeek Harness: per-turn breadcrumb injection via agent/pre-step, a bundled skill provider, and DSH-native task/phase state resolution. No Trellis AGPL code is vendored; workflow semantics are rewritten and licensed MIT.
Gives the DeepSeek Harness agent a configurable humanized whale-girl maid persona (address, self-name, speaking mode), an immersive no-plugin-meta rule, and a lightweight layered memory with query-triggered recall.
DeepSeek Harness (DSH) plugin: a /! command trigger (run one command, analyze output, inspired by Claude Code's ! gesture) and a /terminal command (interactive PTY popup with on-demand history reference).
DSH web client plugin: select text in a chat message and quote it into the composer as an inline reference chip; the full text materializes on submit via the input-trigger reference codec.
Better DSH: archived-session management, task-stop desktop notifications, an update checker, model routing (keyword rules + allowlist-gated model_route), multi-task scheduled triggers (interval heartbeats & fixed-time tasks), and a DeepSeek-style message scroll nav for the dsh web GUI
Host-neutral durable DAG workflows for Agents, Tools, Skills, MCP, triggers, replay, and visual Canvas
DSH skill_manage: agent-managed procedural memory. Create/patch/edit/disable/delete skills in ~/.dsh/skills (user scope) or <projectRoot>/.dsh/skills (project scope), with validation, delete guards (agent-created marker, pin, path confinement), disable/enable via disable-model-invocation, and an English trigger-discipline system prompt section. Built on top of skill-filesystem's hot-reload watcher.
Proactive scheduling v2: value network + file/process/http sensors + HMAC webhook triggers + off-peak window
DeepSeek Harness plugin: a "comic" trigger in the sidebar foot that opens search/download, an importable local library, and an in-modal manga reader. Bundles the jmcomic source so it works without the jmcomic PyPI package.
Message Annotator for DeepSeek Harness: stage numbered quote+comment annotations on assistant messages (selection floating trigger + per-message action), then send them aligned and together to the agent
DSH Web GUI @-mention workspace file picker: input @ to filter git-tracked files plus files re-included via .aiinclude, like Codex/Claude Code file references.
DeepSeek Harness (dsh) plugin: a shell around the dsh-continual-harness self-refinement engine — mounts the engine itself and owns /refine (status / entry browsing / refinement history / one-click rollback / scoped triggers) plus a settings panel; engine still optional at runtime.
Browser audio alert for human-attention events: approval requests, ask-user questions, and finished turns, with per-trigger sounds configurable through the audio-alert settings namespace
Recurring/scheduled tasks (轮询任务/定时任务) plugin for DeepSeek Harness: cron-driven polling jobs as real sessions, natural-language creation, model tools (polling_create/list/edit/delete/trigger) and a web UI. Install-and-go via `dsh plugin add`.
Skill-trigger evaluation: an LLM judge recreates the DSH skill catalog and measures how reliably a skill description routes matching queries.
Configurable near-field trigger that reminds a DeepSeek Harness agent to load the J-Space Cognition Suite only when rules say it is worth it
Strip sandbox_permissions / justification from model tool-call arguments so sandbox escalation is never triggered when the session already has sufficient permission.
Trigger only when the user explicitly invokes $dsh-deliver or clearly asks to hand implementation to DSH. Do not trigger for ordinary implementation, read-only review, commit/push/MR/deploy, or image-only tasks.
DeepSeek Harness plugin to heal third-party gateways that omit the stream terminal event (message_stop/done), which otherwise triggers repeated TRANSPORT retries in the dsh agent loop; scope is configurable per provider and per model.
Trigger GitHub Actions workflow runs and local test pipelines, stream their logs back, and on failure hand the tail of the log to DeepSeek for a Markdown root-cause diagnosis.
Parallel contradiction analysis for DSH conversations: a 0-100 coherence score badge with commentary, manual trigger, and auto-analysis on by default (interval editable, off per conversation or globally). Reuses the conversation's exact prefix so the provider KV cache stays warm.