dsh.so
EN
English中文
deepseek-rlm avatar

deepseek-rlm

Persistent IPython state, durable snapshots, and native recursive subagents for DeepSeek Harness—packaged as a Prime-compatible Cordis plugin bundle.

PluginAI ModelsDeveloperMemoryStorage
Verification
L2 · Structured
Security
Medium
Health
Active
Trust
Silver

What it doesAI

Provides persistent IPython state, durable snapshots, and recursive subagents for DeepSeek Harness as a Cordis plugin.

  • Persistent IPython kernel state across sessions
  • Durable snapshots to save and restore workspace
  • Native recursive subagents for hierarchical AI tasks

AI-generated from the repo README — for reference only.

Installation

dsh plugin --profile web add github:OpenCnid/deepseek-rlm

Install method: GitHub · not yet tested in container (L3+)

Compatibility

DSH VersionStatus
not statedDeclared — not tested

Requirements

  • • Node.js: not stated
  • • DSH: declared "not stated"
  • • External credentials: none detected

Security Report

Automated static scan, not manual review.

DSH.SO AUDITMEDIUM2c9a61
Security reviewAutomated review · daily
MEDIUM

Critical findings in non-blocking categories (dynamic code execution, shell execution, install scripts, obfuscation). Common in CLI/terminal plugins but worth reviewing.

3 critical·0 warning·3 info·15 files scanned
code-exec ×3
Version binding
plugin version0.1.0-preview.0
scanned commit2c9a61c6cd462026-08-20
latest commit2c9a61c6cd46

The scan result is valid for the scanned commit. New commits within 7 days are tolerated (the rating still counts); after 7 days without a rescan the badge shows outdated.

Disclaimer: automated static analysis, not a security guarantee. Always review what you install.

Second-opinion scan (vet · parallel evaluation)
vet: clean

vet verdict: vet: clean · score 94

  • R10 · info [heuristic] — 依赖清单：24 项（@deepseek-ai/cordis, @deepseek-ai/cordis-plugin-loader, @deepseek-ai/dsh-agent, @deepseek-ai/dsh-agent-loop, @deepseek-ai/dsh-agent-loop-testkit, @deepseek-ai/dsh-llm, @deepseek-ai/dsh-jobs, @deepseek-ai/dsh-sandbox, @deepseek-ai/dsh-session, @deepseek-ai/dsh-session-persistence, @deepseek-ai/dsh-session-persistence-jsonl, @deepseek-ai/dsh-session-projection, @deepseek-ai/dsh-subagent, @deepseek-ai/dsh-subagent-spawn-in-process, @deepseek-ai/dsh-system-prompt, @deepseek-ai/dsh-tools, @deepseek-ai/schemastery, @types/node, oxlint, prettier, tsx, typescript, vitest, yaml，供 LLM 审计供应链；已知漏洞核对见后续 OSV 精确版本查询（网络失败静默降级））
  • R3 · info [certain] — 只读 process 成员（能力触达面）：process.env
  • R3 · info [certain] — 只读 process 成员（能力触达面）：process.execPath
  • R3 · info [certain] — 只读 process 成员（能力触达面）：process.stdout
  • R6 · info [heuristic] — 字符串特征：child_process 引用

vet is advisory and does not change dsh’s four-tier level.

Risk: Medium3 critical · 0 warning · 3 info

• Static heuristic scan: done (15 files)

• Dependency vulnerabilities: requires deep audit (L3+)

• Permission sandboxing: requires runtime testing (L4+)

High-risk findings · 3 / 3

  • criticalChild process module usage (Node.js)scripts/pack.ts:1
    import { execFile } from 'node:child_process'
  • criticalChild process module usage (Node.js)scripts/smoke-install.ts:1
    import { execFile } from 'node:child_process'
  • criticalChild process module usage (Node.js)scripts/verify-dsh-patches.ts:1
    import { execFile } from 'node:child_process'

Heuristic static scan — may produce false positives. Review the source yourself before trusting.

Activity

Last commit 2026-08-20 · activity: Active

• Repo created: 2026-08-20

• Stars: ★ 2 · Forks: 0

• Health: Active — committed within last 30 days

Source

GitHub: github.com/OpenCnid/deepseek-rlm

Was this page helpful?