cascadeflow is an in-process runtime for AI agents that makes model and execution decisions during agent loops, including model selection, budget checks, and stop, continue, or escalation actions. It is used to manage cost, latency, quality, policy, compliance, and energy across agent frameworks, while the catalogue skills integrate those decisions into supported agent workflows.
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 lemony-ai/cascadeflow --skill cascadeflow-clawhubgit clone --depth 1 https://github.com/lemony-ai/cascadeflowWrote 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/lemony-ai/cascadeflow/cascadeflow-clawhub)<a href="https://agentmods.dev/skills/lemony-ai/cascadeflow/cascadeflow-clawhub"><img src="https://agentmods.dev/badge/skills/lemony-ai/cascadeflow/cascadeflow-clawhub/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/lemony-ai/cascadeflow/cascadeflow-clawhub"><img src="https://agentmods.dev/badge/skills/lemony-ai/cascadeflow/cascadeflow-clawhub.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00049 | $0.01334 |
| Opus 5.5 | $0.00020 | $0.00534 |
| Sonnet 5.5 | $0.00010 | $0.00267 |
| Haiku 4.5 | $0.00005 | $0.00133 |
Grade A, and why
cascadeflow 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 1mo 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 — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CascadeFlow: Cost + Latency Reduction | 17+ Domain-Aware Models + OpenClaw-Native Events
Use CascadeFlow as an OpenClaw provider to lower cost and latency via cascading. Assign up to 17 domain-specific models (for coding, web search, reasoning, and more), including OpenClaw-native event handling, and cascade between them (small model first, verifier when needed). Keep setup minimal, then verify with one health check and one chat call.
Why Use It
- Reduce spend with drafter/verifier cascading.
- Run 17+ domain-aware model assignments (code, reasoning, web-search, and more).
- Support cascading with streaming and multi-step agent loops.
- Handle OpenClaw-native event/domain signals for smarter model selection.
Security Defaults
- Install from PyPI and verify package artifact before first run.
- Keep the server bound to localhost by default.
- Use explicit auth tokens for chat and stats endpoints (recommended for production).
- Expose remote access only behind TLS/reverse proxy with strong tokens.
- Use least-privilege provider keys (separate test keys from production keys).
How It Works
- OpenClaw sends requests to CascadeFlow through OpenAI-compatible
/v1/chat/completions. - CascadeFlow reads prompt context plus OpenClaw-native event/domain metadata (for example
metadata.method,metadata.event, and channel/category hints). - CascadeFlow selects a domain-aware drafter/verifier pair (small model first).
- If quality passes threshold, drafter answer is returned (cost/latency advantage).
- If quality fails threshold, verifier runs and final answer is upgraded.
- The same cascading behavior is supported for streaming and multi-step agent loops.
Advantages
- Lower average cost by avoiding verifier calls when not needed.
- Lower average latency for simple and medium tasks.
- Better quality on hard tasks through verifier fallback.
- Better operational handling through OpenClaw-native event/domain understanding.
Quick Start
Or ask your OpenClaw agent to set it up for you as an OpenClaw custom provider with OpenClaw-native events and domain understanding.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 116 lines · 49 tokens per session scan A e702015ea1b6
cascadeflow is a skill published in the GitHub repository lemony-ai/cascadeflow (3,936 stars, last pushed 28d ago), licensed MIT. It adds 49 tokens to every session and 1,334 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 2026-08-30.
Other skills, from other repositories
git-memory-remember
Save one durable memory into Tau Git Memory.
git-memory-recall
You are retrieving facts from Tau Git Memory.
html-artifacts
Author the HTML for a plan artifact, dashboard iframe, or Slack attachment — structure, design plan, available runtime, theming, and craft. Read this before writing HTML for saveplan, outputiframe, or slackattachhtml.
hive.slack-notifications-setup
Set up a Slack notification channel (Sentinel) for a colony by driving the browser — reuse or create the "Hive Sentinel" Slack app from a JSON manifest, install it, capture the bot + app tokens, create/select the channel via the Slack API, and turn Sentinel on so the colony can ping the user on Slack and accept…
hive.chart-creation-foundations
Required reading whenever any chart tool is available. Teaches the one-tool embedding contract (call chartrender → live chart appears in chat AND a downloadable PNG lands in the queen session dir), the ECharts (data viz) vs Mermaid (structural diagrams) decision, the BI/financial-grade aesthetic baseline (no…
browser-edge-cases
SOP for debugging browser automation failures on complex websites. Use when browser tools fail on specific sites like LinkedIn, Twitter/X, SPAs, or sites with Shadow DOM.