Plugins

Claude Code-style auto-memory, natively: typed memory files (user/feedback/project/reference) with a MEMORY.md index auto-injected into the system prompt (zero footprint when empty); memory_write/read/list/delete follow Claude Code write rules (dedupe-and-update, absolute dates, cross-links); workspace-scoped and user-scoped layers, cross-process file locking with orphan-lock recovery, plain Markdown on disk, no external services, no embeddings.

Red-team engagement mode: send one target organization name and a planner session runs a five-role execution team (recon, asset triage, vulnerability discovery, exploitation, internal pivot) under a user-configurable concurrency limit (three by default, changed from the console and taking effect immediately), preflight-checking skills and resources and asking once for any missing key or VPS first, guided by a built-in first-run onboarding skill; scoring follows the merged intrusion scoring rules (8 categories / 25 points) with server-side caps, highest-privilege-wins and per-service dedup, and self-registered accounts never score — a login that fits no more specific category is scored as controlling a web application system rather than dropped; findings land in a local SQLite fact base with discovery timestamps, shown in a persistent right-side console of 9 tabs that follows the selected session — switch sessions and it shows that session's target organization, resolving sub-agent sessions to their root session and falling back to the session title, the console footer always naming which of the two matched (its per-session binding file, or the title); (asset mapping with a discovery timeline and the live test view, the agent roster with its concurrency setting and editable role prompts, session/tunnel state, findings — vulnerabilities, credentials and access sessions plus the attack files archived against each vulnerability — five-stage attack chain, scoring targets, an evidence-screenshot capability (register a screenshot per score point, at most three representative shots per point so a bulk result such as dozens of endpoints does not flood the console or the report, viewed inline in the console and embedded in the report, with on-site screenshots kept strictly apart from record-generated evidence cards), and a report that spells out how each score was obtained (actions, exact commands, credential provenance, tunnel build commands, a reproduction entry that is either a paste-ready HTTP request or a runnable command, plus success criteria and an evidence index), a category-organised POC/EXP knowledge base, and a skill library that reports for every unusable skill why it is unusable and the concrete steps to repair it); ships 14 reconnaissance/assessment skills (the npm package deliberately carries no tool auto-installer and no attack-chain skills — those ship as a separate toolkit Release attachment that the operator fetches and runs themselves), 56 redteam_* tools and one-command self-update.

Evidence-first attention supervision for DeepSeek Harness with C0–C3 policy, deduplicated Inbox alerts, quiet hours, and redacted outcome records.

File upload with color-coded attachment cards (session-isolated storage, sha256 dedup, TTL sweep) plus a content-sniffing read_document tool for PDF/DOCX/XLSX/TXT.

Reversible context slimming for DeepSeek Harness: TOON-style lossless tabular re-encoding of uniform JSON arrays, stride-sampled table windows with line numbers, and cross-turn dedup for repeated tool output; every replacement passes a byte-plus-token gate (≤72% bytes, strictly fewer estimated tokens) and the original is spilled to disk behind a `[save-token #id]` notice retrievable in one call via the `save_token_expand` tool, with billed and avoided token metering on a settings dashboard.

Project memory for dsh with native task-system integration: session todos and file reads auto-condense into cross-session task records (task↔file links), so workflows are switchable and resumable without re-contextualizing. Read-time memory (no upfront scan): extracts code symbols, summarizes docs, and cross-links docs↔symbols; optional TS semantic enhancement (auto on when TS installed). Experience notes auto-dedup. Every recall includes a traceable citation. Pure JS, single dep, works out of the box, silent operation, zero overhead.

A self-contained second brain for DeepSeek Harness — auto-captures durable dialogue knowledge into a local Markdown vault (custom dedup, CJK search, git-friendly), with a graphical knowledge base and an interactive knowledge graph (legend filter, time-dimension coloring, box-select, merge, delete, PNG export/import).

Local literature knowledge-base RAG: 10 tools (kb_ingest, kb_search, kb_rag, kb_zotero, kb_dedup, kb_clear, kb_fetch, kb_scope, kb_stats, kb_status), hybrid BM25+vector+reranker search, cited QA with clickable DOI links, all-local bge embeddings over a single-file SQLite index. Needs Python 3.9+ with pymupdf, faiss-cpu and sentence-transformers; the `npx dsh-kb-rag-install` installer sets those up and pre-downloads the model.

Two-way WeChat (iLink) bridge: turn-end and approval-request push, in-chat approve/reject and message injection, persistent dedup and convergent long-reply chunking; channel layer extensible to other IMs.

Auto-log failed tool calls across native tools, PTC run_code, and inline invocations: dedup and count root causes into a skill so repeated mistakes fade.

Human-readable Markdown memory for DeepSeek Harness: user-level (~/.deepseek-harness/MEMORY.md) + workspace-level memory injected every turn, auto daily-log writing with distillation, bridges existing WorkBuddy/CodeBuddy memory dirs, memory_note / memory_note_user / memory_read tools with built-in dedup, and a bilingual settings page. Plain-text storage, no database.

Live skill provider that adds skills from 60+ other agents' directories (Claude Code, Codex, Cursor, Gemini CLI, Copilot), Claude Code plugin marketplaces, and any git repository of skills to the dsh catalog without copying or symlinking; deduplicates, repairs frontmatter, renames name collisions, and ships a CLI (list, add, sync, doctor).

Execution-discipline suite for DSH with identical-retry interception, forced reflection, todo enforcement, memory deduplication, and experience-to-skill promotion (11 plugins + 6 skills).

Codex-style review of session file changes: per-turn cards (optionally taking over the shell edited cards) whose totals are the turn net diff, a review tab with syntax-highlighted diffs that never wrap, real line numbers, open-at-line and a draggable split, a hover preview of the whole file with three lines of context around each change, one-click per-turn revert, and the on-disk record size with a clear button in the review toolbar. Records are sharded per session, with content-addressed snapshot blobs that dedupe and are garbage collected. Light/dark presets follow the DSH theme, typography follows DSH and installed skins (or any font on the machine, with a px stepper), and a settings switch injects a think-in-Simplified-Chinese rule into the global AGENTS.md.

Aggregated web search provider that fans one query out to 11 engines (Exa, Parallel, DuckDuckGo, SearXNG, Brave, Tavily, Serper, Mojeek, DeepSeek, Anthropic, OpenAI) concurrently, URL-dedupes and merges the results. Ships a settings card for toggling backends, storing API keys in the credential store, and tuning concurrency and timeouts.

Proactive multi-model scheduling plugin: multi-source signal intake with dedup and urgency ranking, a four-way execute/defer/dismiss/ask-user decision engine, DAG plan generation with parallel model execution, quality reflection with automatic retry and model switch, long-term memory for task patterns and lessons, and strategy evolution via genetic algorithms with sandboxed evaluation and canary rollout.

Community search, similarity analysis, and install/dedupe/build-it-yourself decisions for the dsh-plugin ecosystem.

Message-level full-text search across DSH sessions via an FTS5 trigram index (user / assistant / steering / tool with per-kind filters), deduplicated results across fork/continued sessions, read-only context preview with auto-scroll to the focused message, one-click jump to the real session, and incremental/full index rebuild with health check; zh/en i18n follows the Host locale service.

Multi-channel IM bridge for DeepSeek Harness: route WeChat (ClawBot/iLink), QQ, and Feishu (Lark) chat messages to a DSH agent and relay its replies back, with per-channel sessions and reply dedup.

Data cleaning and data enrichment for CSV/XLSX/JSON enterprise lists in DeepSeek Harness, including spreadsheet cleaning, deduplication, profiling, optional Qichacha MCP and exports.

Fully local vector memory for DSH: local embeddings, SQLite storage, automatic recall injection, dedup with conflict detection, soft-delete recycle bin, online backup, and optional LLM extraction.

Group installed skills and manage MCP servers from a left-panel UI: filter the model skill catalog by enabled groups (union-deduped) and toggle MCP servers independently.

Three-layer local memory for DeepSeek Harness: L1 snapshot memory (MEMORY.md/USER.md) read on demand through the memory_* tools + L2 searchable knowledge base (facts + SQLite FTS5, Chinese search via sliding-window tokenization) + L3 import from Hermes/Claude Code/Codex/WorkBuddy and from approved Agent Hub documents (memory/global, memory/projects, accepted decisions) with content-hash dedup. v0.2.0 removed every automatic injection: memory enters the context only when the model calls one of the 12 memory_* tools, and data stays on the local drive.

Gitee/GitHub AI employee: @ the bot on an issue (Gitee or GitHub), it clones the repo, implements the fix on the branch you name, opens a PR to that branch, and can auto-merge and auto-close the issue. v1.2 adds security scanning: configure scan repos, audit with built-in/custom prompts, deduplicate, and file new findings as issues.