generate-sandbox-policy
ai-openshell/openshellSecurityCoding & code reviewWriting & content
generate-sandbox-policy is a skill in ai-openshell/openshell, last changed 18 Nov 2026. It can run shell commands, reach the network and name web addresses — read the findings before installing.
Generate sandbox security policies from plain-language requirements and optional REST API documentation. Produces L4 or fine-grained L7 network policies and ordered network middleware configuration. Use for API access rules, middleware host selection, failure behavior, or built-in and operator-run middleware attachment. Trigger keywords - generate policy, create policy, update policy, change policy, sandbox policy, network policy, API policy, security policy, allow API, restrict API, network middleware, supervisor middleware.
At a glance
What it can touch
How much a skill can do, not whether it is risky.
Pattern check
Nothing flagged
That isn't proof the skill is harmless.
Commit 438e07b · 20 Nov 2026
How the check worksPattern check: nothing flagged
That isn't proof the skill is harmless.
QAI Index
We rate every listed skill the same way, from 0 to 100.
A guide to compare by, not a verdict on the skill.
How it's scoredGet an email when its check changes.
Keep it with the other skills you use, installed with one command.
Install
Sign in to see how to install it
Free with Google or an email link, no password needed. Then pick your agent and copy its steps.
What it can touch
Sign in to see the file and line behind each one
The summary at the top stays open to everyone. Signing in is free, with Google or an email link.
Pattern check
Pattern check: nothing flagged · commit 438e07b · 20 Nov 2026
We checked every text file in this skill, at commit 438e07b on 20 Nov 2026, against 45 published patterns such as download-and-run commands, hidden text and credential access. We don't run the skill, we don't read the web pages it points to, and we can't judge everything its instructions might lead an agent to do. In published tests, pattern checks like this one missed most malicious skills. Read the skill before you install it — especially anything it asks you to download or run.
How the check worksReport an incorrect finding
Outside links we did not follow
- https://docs.nvidia.com/openshell/latest/how-it-works/policies/schema
- https://docs.nvidia.com/openshell/latest/extensibility/supervisor-middleware.md
- https://docs.nvidia.com/openshell/latest/how-it-works/policies/overview
- https://docs.nvidia.com/openshell/latest/how-it-works/policies/default-policy
- https://docs.nvidia.com/openshell/latest/how-it-works/policies/schema#matcher-semantics
- https://docs.nvidia.com/openshell/latest/how-it-works/providers/google.md
- https://docs.nvidia.com/openshell/latest/how-it-works/inference
- https://api.github.com
- https://integrate.api.nvidia.com
- http://data-svc.internal:8080
Format check
Each agent reads SKILL.md a little differently. We check the file against each one's published rules; we don't run the skill in any agent.
Agent Skills specPasses with warnings
The format our Agent Skills docs set out, held to the strictest version of each rule among the agents we check.
- Warning
The body is 676 lines; the spec suggests under 500. B1
Fix: Move detailed sections into reference files and link to them from SKILL.md.
- Warning
The body is about 9,616 tokens (approx.); the spec suggests under 5,000. B2
Fix: Move detail into reference files that the agent reads only when needed.
- Note
There is no license field, so readers can't tell whether they may reuse the skill. O2
Fix: Add license: with an SPDX name such as MIT or Apache-2.0, or the name of a bundled licence file.
- Note
Reference files over 100 lines with no table of contents at the top: examples.md (874 lines). B5
Fix: Add a short list of the sections at the top of each.
- Warning
Claude CodePasses
Claude Code reads its own extra fields, ignores unknown ones, and still loads a skill whose frontmatter is broken, without its fields.
- Note
The body is 676 lines; the spec suggests under 500. Claude Code's docs say the same. B1
Fix: Move detailed sections into reference files and link to them from SKILL.md.
- Note
The body is about 9,616 tokens (approx.); the spec suggests under 5,000. After compaction Claude Code keeps only the first 5,000 tokens of a skill. B2
Fix: Move detail into reference files that the agent reads only when needed.
- Note
Reference files over 100 lines with no table of contents at the top: examples.md (874 lines). B5
Fix: Add a short list of the sections at the top of each.
- Note
Claude apps & API uploadPasses
Uploading to claude.ai or the Claude API, or packaging with Anthropic's script, accepts only the six spec fields and stops on anything else.
- Note
The description is 531 characters. The claude.ai Help Center says 200 at most, though Anthropic's API docs and validator allow 1,024. D4
Fix: If the claude.ai upload refuses it, trim the description to 200 characters.
- Note
The body is 676 lines; the spec suggests under 500. Anthropic's guidance says the same. B1
Fix: Move detailed sections into reference files and link to them from SKILL.md.
- Note
The body is about 9,616 tokens (approx.); the spec suggests under 5,000. Anthropic's guidance says the same. B2
Fix: Move detail into reference files that the agent reads only when needed.
- Note
Reference files over 100 lines with no table of contents at the top: examples.md (874 lines). B5
Fix: Add a short list of the sections at the top of each.
- Note
- OpenAI (Codex & API)Passes
- VS Code & CopilotPasses
SKILL.md
Files
2 files
| Path | Size |
|---|---|
| SKILL.md (opens in a new tab) | 38.3 KB |
| examples.md (opens in a new tab) | 24 KB |
Changelog
Newest first: each commit that changed its folder on GitHub, and each change to its listing here.
- Moved here from
/skills/nvidia/openshell/generate-sandbox-policy, which now leads to this page - Updated: fix(policy): align request authorization with route protocol (#5274) (974c2c6 (opens in a new tab))
- Updated: feat(policy): lint authored policy files offline (#5358) (ecce595 (opens in a new tab))
- Updated: feat(middleware): introduce improved HTTP request/response hooks for … (f37687a (opens in a new tab))
- Updated: docs: correct security boundaries and runtime guidance (#5331) (9525bec (opens in a new tab))
- Updated: fix(network): align REST and WebSocket endpoint handling (#5336) (9321d16 (opens in a new tab))
- Updated: fix(sandbox): validate filesystem policy anchors (#5278) (c162ccc (opens in a new tab))
- Listed on QAI
- Updated: feat(supervisor): let operators require specific middleware bindings … (f5d9223 (opens in a new tab))
- Updated: fix(network): bind hostname policy to supervisor DNS (#4602) (c5519ea (opens in a new tab))
- Updated: fix(policy): share endpoint path matching between Rust and Rego (#4336) (b959bb6 (opens in a new tab))
- Updated: fix(vm): reject conflicting workload identity selectors (#4036) (d676e03 (opens in a new tab))
QAI Index badge
Show this skill's QAI Index in a README or on a website. It links back to this page and changes when we refresh the catalog.
[](https://qai.io/skills/ai-openshell/openshell/generate-sandbox-policy)More from ai-openshell/openshell
Other Security skills
Same category, most repo stars first.