Weekly archive Week 44
Week 44: what changed
All 52 tools re-scored on 26 Oct 2026, compared with Week 43.
- Scores changed
- 41
- Label changes
- 2
- Tool releases, 7 days
- 31
Plus 5 tools relabelled to “No dependencies scanned” (scan result unchanged).
- RAGFlownow 71+6
- OpenHandsnow 67+6
- Open WebUInow 72+4
- MarkItDownnow 57+4
- KiroCrewnow 48+4
- OpenCodenow 70−13
- Crawl4AInow 46−6
- Clinenow 64−4
- Codexnow 56−4
- skill-recordernow 25−4
This week's top 10, 3 weeks of history
Score per week from our weekly history files. Yellow line = #1.
- 1FirecrawlW42, W43, W44: 77, 79, 7878No issues found
- 2Gemini CLIW42, W43, W44: 70, 70, 7373Caution
- 3OllamaW42, W43, W44: 71, 74, 7272Caution
- 4Open WebUIW42, W43, W44: 69, 68, 7272Caution
- 5n8nW42, W43, W44: 69, 72, 7272Caution
- 6RAGFlowW42, W43, W44: 64, 65, 7171Caution
- 7OpenCodeW42, W43, W44: 81, 83, 7070Caution
- 8LobeHubW42, W43, W44: 76, 68, 6868Caution
- 9OpenHandsW42, W43, W44: 64, 61, 6767Caution
- 10Claude CodeW42, W43, W44: 65, 65, 6767No dependencies scanned
No tools were added since Week 43.
Top 10 as a table
| # | Tool | Use case | Score | vs Week 43 | Safety | Stars |
|---|---|---|---|---|---|---|
| 1 | Firecrawl | Research & web data | 78 | ▼ 1 | No issues found | 195,395 |
| 2 | Gemini CLI | Coding agents | 73 | ▲ 3 | Caution | 108,389 |
| 3 | Ollama | Local AI | 72 | ▼ 2 | Caution | 184,784 |
| 4 | Open WebUI | Local AI | 72 | ▲ 4 | Caution | 156,767 |
| 5 | n8n | Automation & agent frameworks | 72 | = | Caution | 208,612 |
| 6 | RAGFlow | RAG, knowledge & memory | 71 | ▲ 6 | Caution | 92,871 |
| 7 | OpenCode | Coding agents | 70 | ▼ 13 | Caution | 221,625 |
| 8 | LobeHub | Local AI | 68 | = | Caution | 83,494 |
| 9 | OpenHands | Coding agents | 67 | ▲ 6 | Caution | 91,860 |
| 10 | Claude Code | Coding agents | 67 | ▲ 2 | No dependencies scanned | 153,384 |
Safety label changes
- wshobson/agentsNo issues found changed to Caution6 critical-severity known vulnerabilities in runtime dependencies (OSV)
- MastraNo issues found changed to Caution1 critical-severity known vulnerability in runtime dependencies (OSV)
5 relabels without a scan change
Claude Code, Superpowers, awesome-mcp-servers, smolagents, MarkItDown: “No issues found” became “No dependencies scanned” because these repos have no lockfile to check.
Safety counts this week
36 tools with at least one critical-severity known vulnerability in runtime dependencies (OSV). 7 tools with an agent or skill file that tells the agent to pipe a remote script into a shell. Counts are this week's scan only; since 27 Oct 2026 each critical advisory counts once per package per tool, so totals are not compared with earlier weeks.
31 tool releases in 7 days
By job.
- Coding agents10 / 14
- Agent skills & MCP4 / 9
- Automation & agent frameworks9 / 14
- Research & web data2 / 6
- RAG, knowledge & memory1 / 3
- Local AI4 / 5
- Media & creative1 / 1
All releases and models this week
Tracked tools (31)
- RAGFlow
dev-20261026 - magpie
v0.1.1565 - Cline
desktop-v0.0.49 - AutoGPT
preview-seed-fixture - llama.cpp
b11889 - Firecrawl
v2.11.608 - KiroCrew
v0.10.0 - Pydantic AI
clai2-bleeding - Ollama
v0.40.11 - LobeHub
v0.0.0-nightly.pr21596.36722 - Open WebUI
v0.12.4 - MCP reference servers
v2.0.0 - Gemini CLI
v0.67.0-nightly.20261025.gcd1b45c… - OpenCode
github-v1.2.28 - goose
gdk-v0.1.0-alpha.13 - CrewAI
1.15.30 - LangGraph
1.2.17 - Claude Code
v2.1.310 - Agent Development Kit (ADK, Python)
v2.13.0 - deepseek-harness
dsh-v0.2.1-rc.3 - Codex
rust-v0.165.0-alpha.1 - OpenHands
v1.33.0 - Context7
@upstash/context7-sdk@0.6.0 - n8n
n8n@1.123.87 - Dify
1.18.0 - LangChain
langchain-core==1.7.0 - Remotion
v4.0.538 - Superpowers
v7.0.1 - MarkItDown
v0.1.9 - Mastra
mastracode@0.48.0 - GitHub MCP Server
v2.0.3
Models (1)
- Moonshot AI (Kimi) Kimi K3.1 Pro open weights announced
Get this page by email every week
One email a week: the new top 10, the biggest score changes and every safety label that changed.
Free. Double opt-in: you confirm by email first, unsubscribe in one click.