OmniRoute is an AI gateway that gives coding agents one endpoint for accessing models from many providers, with quota-aware fallback between them. It is for people using Claude Code, Codex, Cursor, OpenCode, Cline, Copilot, and similar tools who want a single route to their available AI models.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add diegosouzapw/OmniRoute --skill green-prsgit clone --depth 1 https://github.com/diegosouzapw/OmniRouteWrote 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.
[](https://agentmods.dev/skills/diegosouzapw/omniroute/green-prs)<a href="https://agentmods.dev/skills/diegosouzapw/omniroute/green-prs"><img src="https://agentmods.dev/badge/skills/diegosouzapw/omniroute/green-prs/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/diegosouzapw/omniroute/green-prs"><img src="https://agentmods.dev/badge/skills/diegosouzapw/omniroute/green-prs.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00051 | $0.00810 |
| Opus 5.5 | $0.00020 | $0.00324 |
| Sonnet 5.5 | $0.00010 | $0.00162 |
| Haiku 4.5 | $0.00005 | $0.00081 |
Grade A, and why
green-prs 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 13d 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.
How it starts
The opening of the file, as written. The whole thing — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Green the selected PR queue
Read ../_shared/base-green.md and ../_shared/validation-gate.md first.
Read ../_shared/protected-surfaces.md before any instruction-surface change or merge.
The current candidate's workflows and versioned policy define required gates;
do not assume release PRs are fast-only or use a handwritten gate count.
1. Resolve scope and ownership
Use the owner's PR list. Otherwise consume eligible merge-ready / fix-in-place
plans in _tasks/pipeline/prs/1-analyzed/; use the whole queue only when requested
or no pipeline items exist. Report drafts separately unless the owner included them.
Record current state, author, head SHA, base branch/SHA, freeze and worktree ownership.
An in-flight worktree belonging to another session means HOLD for that item.
Never work in the shared checkout, stash, or silently retarget a frozen release.
Inspect the exact base first. An incident is diagnostic history, not proof of inheritance. Match each failure signature against the exact base under comparable runtime and configuration. Repair confirmed base defects in a dedicated PR before using them to assess contributor changes. Base-red does not waive a required gate.
2. Prepare and validate
Use one isolated branch/worktree per PR. Before editing a contributor branch, verify the checked-out SHA equals its observed live head. Before pushing, verify the remote still equals the observed predecessor; a concurrent change means HOLD. Keep the original author's work and attribution, add appropriate co-authorship, and never overwrite or close the PR to take credit. A locked fork needs the owner-reviewed credit-preserving fallback, not a destructive reconstruction.
Run a RED reproduction, implement the proven fix, then focused and affected suites. Dependencies must match the candidate lockfile/toolchain; borrowed tools are diagnostic only. A build needs physical dependencies, not an external symlink. Preserve raw logs, exits, artifacts, test discovery and all existing assertions.
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.
- 13d ago First seen · 66 lines · 51 tokens per session scan A 019c5e3e64e5
green-prs is a skill published in the GitHub repository diegosouzapw/OmniRoute (77,028 stars, last pushed today), licensed MIT. It adds 51 tokens to every session and 810 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-10-03.
Other skills, from other repositories
ln-63-deployment-engineer
Prepares CI/CD and remote infrastructure, then executes authorized rollouts with health and recovery checks; not standalone local container packaging.
pr-babysitter
Monitors or repairs an open GitHub PR: CI failures, conflicts, review threads, and merge readiness, reporting state changes. Use when asked to "watch this PR", "fix CI", "resolve conflicts", or "address review comments".
verification-quality
Truth scoring, code-quality verification, and automatic rollback with a high accuracy threshold. Use when validating agent output before it ships, gating changes on quality, or adding a verify-and-rollback safety loop.
loki-mode
An autonomous software factory that knows what it is supposed to deliver, and proves it did. Use when the user says Loki Mode or asks to build, fix or verify software autonomously.
checkpoint-mode
Pause for review every N tasks - selective autonomy pattern.
loki-mode
Launch Loki Mode autonomous SDLC agent. Handles PRD-to-deployment with minimal human intervention. Invoke for multi-phase development tasks, bug fixing campaigns, or full product builds.