Plugins

Runs local security audits of AI API relays and LLM proxies from DeepSeek Harness, producing Markdown reports for prompt injection, model substitution signals, tool-call rewriting, error leakage, stream integrity, and profile-gated Web3 risks.

Risk-gated approval automation for DeepSeek Harness: flash pre-classifies whether a write/command is irreversible — safe operations are auto-approved, dangerous ones are escalated to human approval (fail-safe). File-diff review with one-click revert and session-scoped snapshots (v0.5.1: precise snapshots via tool-call parameter tracing, incl. human-approval cases).

DSH plugin: Claude-style Chinese risk control & conversation autonomy for DeepSeek Harness web

Quantitative R&D toolkit for DeepSeek Harness — 59 tools across six domains covering market data, indicators, factor evaluation, walk-forward ML validation, risk (VaR/CVaR/drawdown/Beta with Kupiec POF), options, bonds, FICC and fund simulation, with an end-to-end PDAT→PET research pipeline.

Research-only trading workbench bundle: typed market-data provider seam with a CSV reference provider, multi-timeframe indicator regime snapshot and OHLCV tools, interactive chart cards in dsh web with provenance-gated model annotations, and a risk-guard denying execution-shaped tool calls at tools/pre-execute.

QA pipeline as 10 agent skills: requirement analysis, test strategy, case writing and review, E2E (Playwright) and API automation, exploratory, regression scope and bug analysis, backed by a shared knowledge base with format rules, risk model and schema validators.

Delete DSH sessions from the web UI and desktop client: header danger button and session-row menu item with a risk-consent dialog; host endpoint and agent tool remove the session log, projection cache, and workspace accounting.

Classifies pending tool calls with Jev (TypeSafe AI's non-generative decision model) on tools/pre-execute — confident high-risk calls are denied, ambiguous ones escalated to approval — and auto-approves clearly-granted reversible calls on approval/request behind argument-evidence gating. Also subclasses the session-reference resolver so snapshots keep Jev-scored messages instead of dropping oldest-first. Degrades to stock behavior on any provider failure; shadow mode with a /jev-stats command; TypeSafe or OpenRouter endpoints; 64 tests.

White-box causal guardrail for dsh: every tool call is adjudicated before execution along a deterministic causal logic chain — blocks destructive and credential-file operations, escalates repeated boundary violations, cuts faulting links until verified-fixed, and gives risk verdicts on un-auditable execution; runs fully local with zero API cost, plus 6 white-box self-check tools.

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

Expert-mode agent preset for DeepSeek Harness (v0.9.2, npm dsh-expert-mode, bilingual EN/ZH): a chief coordinator plus 17 domain-expert subagents with automatic task delegation. Features: taskboard scheduler (file-system task state machine pending/ready/running/done/failed, dependency DAG, atomic claim, retry, crash recovery), quality gates (5-stage pipeline for high-risk tasks: requirement clarity, implementation, verification, independent review, integration, with 2-round rework limit), Five-Anchor constraint (review/convergence/anti-drift/collaboration-check/resource-awareness per turn), Near-distance Guidance (identity/task/output template per expert), progressive disclosure (~28% token savings), expert persistence, inter-expert file message bus (direct P2P comm, zero coordinator relay), cross review, experience pool, fast-track for simple tasks, fault recovery with auto-retry. Experts: data analyst, copywriter, legal review, product manager, frontend, UI/UX, architect, social media ops, growth hacker, quant finance, finance, backend, DevOps, database, QA, security.

LLM-assisted auto approval with countdown fallback for the Auto permission preset: static rules, risk tiers, breaker and file audit.

A second brain for your DeepSeek Harness agent — strong-model review before risky tool calls, plus 67 models from one wallet. No accounts, no API keys.

Tool-call discipline for DeepSeek Harness: injects a static behavior-constraint prompt section, drops the parallel pool to 1 after a tool failure (restoring after 3 consecutive successes), nudges the model to stop retrying after repeated failures, verifies writes with .bak rollback, gates high-risk commands, and tracks compliance stats via /behavior-status.

Live security beacon for DSH that follows tool activity. Click to view call statistics, risk provenance chains and blocked actions, with read-only task restrictions and tool metadata change detection.

Browse and install DSH plugins without leaving the harness: a /store command, a Settings → Plugins tab, two agent-callable tools and a bundled skill, with an approval gate that lists the detected risk flags before anything installs.

Move a session to another folder from the Web UI via drag & drop or a menu picker, permanently delete a session with a risk-consent dialog, and AI-rename a session by summarizing the whole conversation (typo-aware). Each action also ships as an agent tool.

Financial research workflow and portfolio risk tools with source discipline for current market facts.

DeepSeek Harness plugin that blocks risky npm installs with metadata and deep supply-chain scans

Claim-verification skill that checks design documents and plans against codebase facts, escalating behavioral claims to executable logic-primitive verification for state machines and data models, with before/after regression, domain constraints, and concurrency risk mining.

Static security scanner for DSH plugins: read-only audit (exec, credentials, exfiltration, obfuscation, install scripts, bundle manifest) with a 0-100 risk score.

Finance data toolkit for DSH agents: number and currency formatting (incl. Chinese wan/yi units), return and CAGR math, valuation and profitability ratios, PV/FV, risk metrics, and a data-quality checklist.

A content-aware PDF reading plugin for vision models: profiles each page for figures (vector and raster), tables, formula risk and double-column layout, then applies content-aware hybrid extraction, rendering figure/table/formula pages as high-DPI region crops. Provides a low-resolution preview to understand the page layout, and renders a specified region at high resolution. Packaged as multiple tools for agents.

A-share multi-agent analysis behind /trading-agent, ported from TradingAgents-AShare, running 14 roles (analysts, bull/bear debate, risk review, trader) through the harness LLM route.