Plugins

Open-source CMA-compatible agent runtime for any model, with MCP tools, sandboxed sessions, audit, replay, and a local console. Includes a native DeepSeek Harness bundle over stdio MCP.

Scheduled coding runs in fresh agent sessions with auditable history.

Local security audit: config, plugin origins, sessions, network exposure — read-only redacted risk report.

Context injection audit: token costs of instruction chains / skill catalogs / tool schemas, duplicate and conflict detection.

Portable Agent Skill for developing and auditing DeepSeek Harness plugins, with an optional profile-installable DSH bundle adapter.

Adds a built-in plugin catalog to DSH Web with search, status explanations, and safe toggles for audited UI plugins.

Appends local hash-chained JSONL receipts for tool results and session events without storing prompts, tool arguments, result text, or raw session IDs.

Continual self-evolution: versioned, auditable, rollback-safe harness state (prompts, memory, skills, subagent specs) refined from session trajectories, with review gates and hot-reloaded skills.

Login gateway for the DSH web UI: password door with first-run setup, bcrypt + at-rest encryption (AES-256-GCM/HMAC), brute-force lockout, audit log, TLS 1.2+ with 80→443 redirect, CSRF, anti-framing.

Research agent suite: long-horizon goal control with auditable evidence chains, cron scheduling, multi-agent expert teams, self-evolving research memory (FTS5 + RRF recall), project workspaces, and a custom workspace UI.

Evidence-first operating kit: six read-only tools (capability catalog, staged workflow plans, packaged skill reader, bounded local memory search, repository release audit, release checklist) and five packaged skills - plans and audits, never remote writes.

Session execution analytics: steps, tool calls, failures, repeated actions, token usage and verification signals, rendered as text/Markdown/JSON reports.

Bounded, layered, approval-gated, auditable cross-session memory: a typed `ctx.memory` seam with a zero-dependency SQLite provider, a `memory` tool, and frozen snapshot injection; every write passes the approval gate and stays reconstructable from the session log.

Declarative Claude Code-style permission rules for DeepSeek Harness: ordered allow/deny/ask rules with tool-name, argument (glob/regex), and workspace-path matching on the tools/pre-execute waterfall, full session-log audit, and HMR rule reload.

Engineering-discipline guard: grill the requirements before the first edit, enforce red/green test evidence gates, and audit the delivery with a forked adversary (grill-requirements skill + tool-policy gates).

Build auditable KB packs (SQLite FTS5) from md/txt/docx/pdf with deterministic retrieval and original-text reading.

Bounded local memory with CBDC authority gating: SQLite + FTS5 claims, scoped recall with explainable use/verify/ignore decisions and a full audit trail, /memory commands, ≤3-claim/1200-char injection per call, no extra model call.

Optional provider plugin for dsh-skill-pack-security: registers the pack's skills/ (zh) or skills-en/ (en) edition on ctx.skills. Ships both editions embedded in pack/.

Harness 事件流审计面板：观察事件类型、分发模式、计数与最近事件，帮助插件作者理解 harness 内部发生了什么

The `dsh pm` plugin manager: multi-source search (awesome list + GitHub + npm), install/remove/update per profile, and a doctor audit of manifests, bundle patches, and version drift.

Read-only agent-fleet credential hygiene audit: credential-file permissions, embedded credentials in git remotes (masked in output), and provider token literal counts; zero-dependency and deterministic.

将DeepSeek Harness插件列表按来源分组展示——官方插件与自行安装插件，一目了然。