What it doesAI
Autonomous research agent that iteratively modifies code, trains for 5 minutes, evaluates val_bpb, and keeps or discards changes for ~100 experiments per night.
- Iterative loop: modify code, train 5 min, evaluate val_bpb
- Single-file modification with lightweight program.md skill
- Runs ~100 experiments per night autonomously
AI-generated from the repo README — for reference only.
Installation
dsh plugin --profile web add github:satan9394/dsh-autonomous-researchInstall method: GitHub · not yet tested in container (L3+)
Compatibility
| DSH Version | Status |
|---|---|
| not stated | Declared — not tested |
Requirements
- • Node.js: not stated
- • DSH: declared "not stated"
- • External credentials: none detected
Security Report
Automated static scan, not manual review.
No critical or warning findings in the automated scan. The result is valid only for the commit hash shown.
0.1.0dsh manifest2d8bc5589aa52026-08-202d8bc5589aa5The 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 New commits within 7 days are tolerated (the rating still counts); after 7 days without a rescan the badge shows outdated.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.
vet verdict: vet: clean · score 94
- R10 · info [heuristic] — 依赖清单：1 项（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））
- R12 · info [heuristic] (downgraded) — engines.node=>=20.0.0 低于 DSH 运行要求（>=22.19），可能不兼容
- R9 · medium [likely] — 正则嵌套量词（ReDoS 风险：(a+)+ 类指数回溯）
vet is advisory and does not change dsh’s four-tier level.
• Static heuristic scan: done (3 files)
• Dependency vulnerabilities: requires deep audit (L3+)
• Permission sandboxing: requires runtime testing (L4+)
Activity
Last commit 2026-08-20 · activity: Active
• Repo created: 2026-08-20
• Stars: ★ 0 · Forks: 0
• Health: Active — committed within last 30 days