llm-cli-gateway: Skill for Claude Code

.agents/skills/secure-orchestration/SKILL.md

secure-orchestration is a skill for Claude Code from verivus-oss/llm-cli-gateway. It costs 38 tokens per session (2,815 once invoked), scanned A, original, MIT.

A guide for handling sensitive work with language models through an approval boundary and an evidence-aware audit process. It covers privileged operations, autonomous changes, and security reviews.

In plain words
What is it for?
It is for orchestrating sensitive code changes and security reviews, including approval policies, provider-specific controls, and evidence collection.
Why use it?
It helps ensure security-sensitive requests use the required gateway review path and permission controls.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: installed under .agents/ (shared by several agents); mentions Codex.

This is verivus-oss/llm-cli-gateway's own configuration. It tells Claude Code how to work on llm-cli-gateway itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything llm-cli-gateway configures →

Part of the llm-gateway plugin — 24 skills, 4 commands, 2 MCP servers shipped together

Reuse

Borrowing it

Nothing to install: this file belongs to verivus-oss/llm-cli-gateway. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/verivus-oss/llm-cli-gateway/main/.agents/skills/secure-orchestration/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/verivus-oss/llm-cli-gateway

Made for: Claude Code.

Or install llm-gateway, the plugin that ships this one along with the rest of its 24 skills, 4 commands, 2 MCP servers.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for secure-orchestration

README.md
[![agentmods](https://agentmods.dev/badge/skills/verivus-oss/llm-cli-gateway/secure-orchestration.svg)](https://agentmods.dev/skills/verivus-oss/llm-cli-gateway/secure-orchestration)
Your own site
<a href="https://agentmods.dev/skills/verivus-oss/llm-cli-gateway/secure-orchestration"><img src="https://agentmods.dev/badge/skills/verivus-oss/llm-cli-gateway/secure-orchestration.svg" alt="Measured on agentmods" height="20"></a>
Per session 38 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,815 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00038 $0.02815
Opus 5 $0.00019 $0.01407
Sonnet 5 $0.00008 $0.00563
Haiku 4.5 $0.00004 $0.00281

Measured 8d ago against content hash 80a614803163, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

secure-orchestration scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.agents/skills/secure-orchestration/SKILL.md · 259 lines

How it starts

The opening of the file, as written. The whole thing — 259 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Secure Orchestration

Use this skill for sensitive code, privileged operations, autonomous changes, and security review. A security review request goes through the installed local stdio gateway MCP surface, never a direct provider binary, SDK, connector/shadow gateway, or shell fallback. If that stdio surface is unavailable, repair it or report the review incomplete.

The approval boundary

approvalStrategy:"mcp_managed" is an enforcement boundary only for Claude. The gateway creates a request-scoped strict MCP configuration from provisioned gateway-owned local definitions. Codex, Gemini, Grok, Mistral, Devin, and Cursor must use approvalStrategy:"legacy"; they reject managed approval before launch because their ambient MCP configuration cannot be isolated.

For Claude managed requests:

  • approvalPolicy may be strict, balanced, or permissive.
  • The default policy is balanced; thresholds are strict 2, balanced 5, and permissive 7.
  • Full permission bypasses and unverified execution posture are denied by default. They require the caller's explicit request, an approval decision, and LLM_GATEWAY_APPROVAL_ALLOW_BYPASS=1.
  • Native continuation/fork, workingDir, tool selectors, settings/plugins, additional directories, prompt-file controls, and other posture changes can require that same decision and operator setting. Do not add them casually.
  • Under managed approval, only gateway-owned local definitions explicitly provisioned for the request are eligible. Dynamic package execution, ambient PATH, and provider-configuration overrides cannot bypass that boundary.

Do not imply that gateway-managed approval protects a non-Claude legacy request. Its sandbox, tool, permission, MCP, session, and ACP controls belong to the provider. Query provider_tool_capabilities({cli:"..."}) before relying on any of them.

Safe request patterns

Routine Claude managed inspection:

claude_request({
  prompt:"Audit the supplied evidence packet for security defects. Inspect available source directly and cite evidence.",
  approvalStrategy:"mcp_managed",
  approvalPolicy:"strict"
})

Read the full file on GitHub · 259 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago First seen · 259 lines · 38 tokens per session scan A 80a614803163

Subscribe to this mod's changes

secure-orchestration is a skill published in the GitHub repository verivus-oss/llm-cli-gateway (15 stars, last pushed today), licensed MIT. It adds 38 tokens to every session and 2,815 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.