DeepSeek Harness PluginManifest Valid

dsh-rigorquant

Session-scoped multi-agent quantitative and mathematical research bundle for DeepSeek Harness with isolated lanes and validation gates.

Developer ToolsSecurity & PolicySkills & Workflows

6GitHub Stars1ForksUpdated2026-10-01

Source

Install

$ dsh plugin --profile web add dsh-rigorquant

Plugin Overview & Capabilities

AI-assisted organization based on the public repository snapshot. The content must be grounded in source evidence and does not replace compatibility or security verification.

source-grounded

dsh-rigorquant is an agent preset and plugin bundle for DeepSeek Harness designed for empirical and computational mathematics, economics, and quantitative finance research. It structures a Harness session into a hub-and-spoke multi-agent environment featuring specialized personas (Orchestrator, Explorers, OffGridThinker, DoubleChecker, Adversaries, and Validators). It enforces independent derivations, counterexample hunting, a four-part verification battery prior to numerical implementation, and meta-validation against audit logs.

Key Capabilities

Context-isolated multi-agent team structure adhering to a strict hub-and-spoke communication topology
Independent double-checking from first principles and counterexample-only elimination
Four-part pre-implementation battery verifying closed-form equality, exact invariants, analytic bounds, and statistical hardening
OffGridThinker lane with isolated compute tools (SymPy, NumPy, mpmath, Lean checkers) without web or peer context
Meta-validator (rq_check.py) checking audit records and deliverables for completeness and integrity
Role-based model routing (rq-model-router) supporting fallback strategies across DeepSeek models
Pinned uv-managed compute environment and optional jacobian MCP escalation lane

Useful For

Rigorous quantitative finance modeling, portfolio optimization, and market simulations
Computational mathematics and symbolic derivation validation without hallucination leaks
Academic economics research requiring verifiable derivations and audit-trail logging

Who It Fits

Quantitative researchers and financial engineers using DeepSeek Harness
Computational economists and applied mathematicians
AI researchers conducting automated formal verification and counterexample exploration

Documented Limitations

Requires DeepSeek Harness >=0.2.0-rc.2 <0.2.1 and Agent Teams capability
Unattended execution is scoped only within a single live session and disarms upon session restart
Requires approximately 1 GB disk allocation per active study for the isolated uv environment and caches

DSH Compatibility

Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.

Not tested yetNo runtime compatibility tests have been published yet.

Security Signals

Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.

Dependency Counts

package.json declares 1 runtime, 0 development, 1 peer, and 0 optional dependencies.

info

Dsh Bundle Declared

package.json declares DSH bundle metadata.

info

License Declared

GitHub reports the repository license as MIT.

info

Package Manifest Available

A root package.json was captured and can be inspected by the registry.

info

Source Available

Public GitHub source metadata is available for this registry snapshot.

info

Source & Registry Notes

Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.

Source repository: linxichen/dsh-rigorquant
Registry source: GitHub · dsh-plugin topic
Registry classification: Plugin Bundle
Source checked: 7ec633e · 2026-10-21

This project is independently indexed from public source information. DSH Plugin is not affiliated with DeepSeek or the plugin author. Always check the author repository before installation.