openshell-policy-advisor
Skill19k starsChanged 53 days agoMITView on GitHub
Use when an OpenShell sandbox returns policy_denied, mentions policy.local, or needs a narrow network policy proposal.
1 other project uses a version of this.
Adopted by
Steady: 0 new projects took it in the last 4 weeks, none the 4 weeks before.thin data
Most recently taken up by
- NVIDIA/OpenShellword for word, first seen Oct 13, 2026
Who's talking about anthropics/claude-quickstarts
- Cole Medin“I Forced Claude to Code for 24 Hours NONSTOP, Here's What Happened”YouTube · Dec 4, 2025 · 39k views
- Ray Fernando“Claude SDK: 24-Hour Coding Agent | Cole Medin”YouTube · Dec 8, 2025 · 18k views
- Web Dev Cody“I Let Claude Code Build an App for 24 Hours”YouTube · Dec 8, 2025 · 13k views
- 橘鸦Juya“Claude Haiku 5.5 发布;GPT-6 上线 Chat;Sonnet 降价;Claude Max 获赠API额度【AI 早报 2026-10-08】”YouTube · Oct 8, 2026 · 5.7k views
- DataSense“𝗔𝗜 𝗧𝗵𝗮𝘁 𝗖𝗼𝗱𝗲𝘀 𝗜𝘁𝘀𝗲𝗹𝗳? 𝗥𝗲𝗮𝗰𝘁 𝗟𝗼𝗼𝗽 𝘃𝘀 𝗥𝗮𝗹𝗽𝗵 𝗟𝗼𝗼𝗽 𝗘𝘅𝗽𝗹𝗮𝗶𝗻𝗲𝗱”YouTube · Feb 28, 2026 · 694 views
The biggest 5 of 48 mentions linking to the repository that our crawl found. Each line is their own headline.
See all 48
Talked about on GitHub, YouTube, the web and Bluesky (48 times).
On GitHub · 24
- Nov 7, 2026Add Claude Quickstarts (Computer Use Demo) (anthropics/claude-quickstarts) — reacher-z/Awesome-CUA0 reactions
- Nov 7, 2026Curate CUA resources batch w1-033 (Claude Quickstarts (Computer Use Demo), dsh-antibrow, +3 more) — reacher-z/Awesome-CUA0 reactions
- Nov 4, 2026Checkpointing for unplanned interruptions: extend acp-dispatch --checkpoint (research, plan, open decisions) — djbclark/djbclark-ade0 reactions · 2 comments
- Nov 2, 2026[Feature] runtime-e2e-test: setup mode that detects, proves, and fills project-runtime.md — phananhtuan09/ai-agent-workflow0 reactions
- Oct 31, 2026docs(playbooks): re-anchor skill-authoring and fable-5 verification to the claude.dev blog posts — melodic-software/claude-code-plugins0 reactions · 4 comments
19 more on GitHub
- Oct 17, 2026Run agent task evals against previews — totalolage/deskohub-site0 reactions · 1 comment
- Sep 25, 2026agent-skills.github.anthropics.claude-quickstarts: init at 0-unstable-2026-08-21 — sudosubin/agents.nix0 reactions
- Sep 23, 2026docs(managed_agents): move the self-hosted sandbox variants to claude-quickstarts — anthropics/claude-cookbooks0 reactions · 2 comments
- Sep 21, 2026docs(managed_agents): move the Linear, road trip planner, and MCP server apps to claude-quickstarts — anthropics/claude-cookbooks0 reactions · 2 comments
- Sep 16, 2026docs(managed_agents): move the Slack and Sentry apps to claude-quickstarts — anthropics/claude-cookbooks0 reactions
- Sep 16, 2026docs(managed_agents): move the Slack and Sentry apps to claude-quickstarts — anthropics/claude-cookbooks0 reactions · 2 comments
- Sep 15, 2026Factory.ai 深度调研报告 — Pines-Cheng/blog0 reactions
- Sep 13, 2026/automations: real product widgets + trademark disclaimer — clancybooher/peak-view-windows0 reactions · 1 comment
- Aug 2, 2026goal-maker/bhag: adopt Anthropic's long-run harness (immutable feature_list.json + append-only progress file + fixed bootstrap drill) — jackneil/claude-jacked0 reactions
- Feb 17, 2026Closing the loop - Infrastructure for Agentic Engineering — schemalabz/opencouncil0 reactions
- Feb 3, 2026[Feature] Add Session Persistence, Security Model, and Progress Tracking (v2.2.0) — jnPiyush/AgentX0 reactions · 4 comments
- Jan 13, 2026feat: Add support for ANTHROPIC_API_KEY authentication — nirmalarya/claude-harness0 reactions
- Jan 2, 2026Meta-Library Architecture: Transform Any AI Workflow Knowledge into Executable FSMs — JulesGosnell/claij0 reactions
- Jan 2, 2026Learn from Anthropic's Long-Running Harness Pattern — JulesGosnell/claij0 reactions
- Jan 1, 2026Does this work with Anthropics new Harness for long running agent strategy? — rubycatco/claude-agent-sdk-ruby0 reactions · 2 comments
- Dec 23, 2025Add a think tool — byronxlg/polyagent0 reactions
- Dec 11, 2025feat: Add Autonomous Claude Code Builder system — krzemienski/shannon-framework0 reactions · 2 comments
- Dec 6, 2025Clone Anthropic computer-use-demo into repository — matheus-rech/comp-use0 reactions · 2 comments
- Nov 10, 2025File Editor Tool — strands-agents/sdk-typescript0 reactions · 7 comments
On YouTube · 20
- Dec 4, 2025I Forced Claude to Code for 24 Hours NONSTOP, Here's What Happened — Cole Medin39k views · 198 comments
- Dec 8, 2025Claude SDK: 24-Hour Coding Agent | Cole Medin — Ray Fernando18k views · 36 comments
- Dec 8, 2025I Let Claude Code Build an App for 24 Hours — Web Dev Cody13k views · 109 comments
- Oct 8, 2026Claude Haiku 5.5 发布;GPT-6 上线 Chat;Sonnet 降价;Claude Max 获赠API额度【AI 早报 2026-10-08】 — 橘鸦Juya5.7k views · 1 comment
- Feb 28, 2026𝗔𝗜 𝗧𝗵𝗮𝘁 𝗖𝗼𝗱𝗲𝘀 𝗜𝘁𝘀𝗲𝗹𝗳? 𝗥𝗲𝗮𝗰𝘁 𝗟𝗼𝗼𝗽 𝘃𝘀 𝗥𝗮𝗹𝗽𝗵 𝗟𝗼𝗼𝗽 𝗘𝘅𝗽𝗹𝗮𝗶𝗻𝗲𝗱 — DataSense694 views
15 more on YouTube
- May 2, 2026Anthropic Open-Sourced 5 Repos (230k Stars). You've Used Claude Daily and Cloned None of Them. — Willm601 views
- Apr 17, 2026Spec Driven Discovery — Kai Hendry254 views
- Mar 21, 2026Claude Quickstarts 2026: Computer Use, Coding Agents, and Financial RAG — Blueprint Bytes189 views
- Oct 18, 2026Claude-quickstarts: build AI chat assistants from support articles — Next Horizon103 views · 1 comment
- Dec 7, 2025GitHub - anthropics/claude-quickstarts: A collection of projects designed to help developers quic... — GitHub Daily Trend AI Podcast84 views
- Oct 11, 2026#9 Anthropic’s Initializer + Coding Agent Pattern: Long-Running Agent Harness Breakdown — Vishal81 views
- Aug 8, 202619 Free Anthropic Resources Repo on GITUB that Put You Ahead of 95% of Claude Users — Ai Therapy Labsays it tested it72 views
- Oct 11, 2026Anthropicの定期実行AIエージェント、許可する人なしでどう取りこぼしを防ぐ?|返事を見てから帳簿へ — エンジニアになりたいずんだもん66 views
- Dec 7, 2025anthropics/claude-quickstarts - Gource visualisation — Gourcer58 views
- Oct 11, 2026Why Scheduled AI Agents Break (And How Anthropic Fixed It) — PromptlyAshwin56 views · 1 comment
- Aug 9, 2026Claude 19 Free Resources Repo on GITUB that Put You Ahead of 95% of Claude Users — Ai Therapy Labsays it tested it35 views
- Aug 8, 202619 Free Claude Resources That Put You Ahead of 95% of Claude Users — Ai Therapy Labsays it tested it23 views
- Oct 19, 2026고객 문의 메일, AI 에이전트가 답장 초안까지 써 두게 하는 법 | AX 설계 — 수달빌더22 views · 2 comments
- Oct 9, 2026Claude Haiku 5.5 in GitHub Copilot | AI Daily Brief | 2026-10-08 — Insight AI19 views
- Oct 9, 2026Claude Managed Agents - Build a Scheduled AI Agent From Scratch — ProAI Clips17 views
On the web · 3
- Jul 1, 202630 FREE Claude Skills - by aiwithram - AI With Ram — growtechie.substack.com
- Apr 14, 2026AI Engineer Europe 2026: Takeaways from London — dabase.com
- Jan 31, 2026Anthropic Claude 3 Roadmap for 2026 — linkedin.com
On Bluesky · 1
- Oct 18, 2026🚀 Skyrocketing! 🚀 (200+ new stars) 📦 anthropics / claude-quickstarts ⭐ 18,532 (+453) 🗒 TypeScript A collection of projects designed to… — @github-trending-js.bsky.social2 likes and reposts · 1 reply
Other versions1 project uses a version of this.
Word for word · 1
- NVIDIA/OpenShell★ 21k100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository24 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- managed-agents/assistant-ui/CLAUDE.mdCLAUDE.md
- managed-agents/chat-sdk/CLAUDE.mdCLAUDE.md
- managed-agents/copilot-kit-ag-ui/CLAUDE.mdCLAUDE.md
- managed-agents/daily-brief/CLAUDE.mdCLAUDE.md
- managed-agents/knowledge-wiki/CLAUDE.mdCLAUDE.md
- managed-agents/linear/CLAUDE.mdCLAUDE.md
- managed-agents/mcp-server-typescript/CLAUDE.mdCLAUDE.md
- managed-agents/roadtrip-planner/CLAUDE.mdCLAUDE.md
- managed-agents/self-hosted-sandboxes/CLAUDE.mdCLAUDE.md
- managed-agents/sentry-autofix/CLAUDE.mdCLAUDE.md
- managed-agents/sentry/CLAUDE.mdCLAUDE.md
- managed-agents/slack/CLAUDE.mdCLAUDE.md
- first-runSkill
- verifySkill
- assistant-uiSkill
- chat-sdkSkill
- verifySkill
- knowledge-wikiSkill
- linearSkill
- mcp-server-typescriptSkill
- roadtrip-plannerSkill
- sentrySkill
- slackSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 8
When a request fails with policy_denied, do not treat the denial as final if the user task still needs the request.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
OpenShell Policy Advisor
When a request fails with policy_denied, do not treat the denial as final if the user task still needs the request. Read /etc/openshell/skills/policy_advisor.md, inspect http://policy.local/v1/policy/current, submit the narrowest proposal to http://policy.local/v1/proposals, wait on /v1/proposals/{chunk_id}/wait?timeout=300, and retry only after approval with policy_reloaded: true.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

