dsh.so
EN
English中文
dsh-evolve avatar

dsh-evolve

Evidence-driven execution experience plugin for DeepSeek Harness

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

What it doesAI

Records repeated failures, corrections, and success paths for DSH agents, distilling local experiences for temporary retrieval in future tasks.

  • Detects repeated errors, tool calls, and no-progress traces deterministically
  • Background sidecar mines corrections, successful procedures, failure patterns
  • Retrieves max 3 relevant experiences; not persisted as resident context

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

Installation

dsh plugin --profile web add github:Atman-Angle/dsh-evolve

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 AUDITMEDIUM91d3f0
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.

1 critical·0 warning·0 info·15 files scanned
code-exec ×1
Version binding
plugin version0.1.0dsh manifest
scanned commit91d3f0a73b1a2026-08-20
latest commit91d3f0a73b1a

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 New commits within 7 days are tolerated (the rating still counts); after 7 days without a rescan the badge shows outdated.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 81

  • R10 · info [heuristic] — 依赖清单：14 项（@deepseek-ai/cordis, @deepseek-ai/dsh-agent, @deepseek-ai/dsh-llm, @deepseek-ai/dsh-session, @deepseek-ai/dsh-tools, @deepseek-ai/schemastery, @types/node, typescript, vitest, @deepseek-ai/cordis, @deepseek-ai/dsh-agent, @deepseek-ai/dsh-llm, @deepseek-ai/dsh-session, @deepseek-ai/schemastery，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））
  • R12 · info [certain] (downgraded) — 入口文件缺失：./lib/index.js
  • R2 · medium [likely] — require() 动态模块加载（npm 包内能力触达）
  • R3 · info [certain] — 只读 process 成员（能力触达面）：process.stderr
  • R3 · info [certain] — 只读 process 成员（能力触达面）：process.stderr

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

Risk: Medium1 critical · 0 warning · 0 info

• Static heuristic scan: done (15 files)

• Dependency vulnerabilities: requires deep audit (L3+)

• Permission sandboxing: requires runtime testing (L4+)

High-risk findings · 1 / 1

  • criticalChild process module usage (Node.js)benchmarks/runtime-overhead.ts:13
    import { spawnSync } 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: ★ 0 · Forks: 0

• Health: Active — committed within last 30 days

Source

GitHub: github.com/Atman-Angle/dsh-evolve

Was this page helpful?