Category 2 infrastructure demo

Fiber DevKit

A Rust CLI for local Fiber Network orchestration, repeatable payment scenarios, structured failure diagnosis, route prediction, report artifacts, and CI smoke tests.

Watch it work Read the README
Fiber DevKit architecture A Rust CLI controls local Fiber nodes, scenario YAML files, diagnostics, route prediction, and CI reports. fiber CLI init · up · run · doctor Scenario YAML payments, failures, asserts Report Artifacts report.md · trace.json · logs node-1 hub · FNN RPC node-2 leaf · channel node-3 leaf · channel Taxonomy FIBER_LIQ_001 etc.

Flow Maps

These diagrams explain how DevKit moves from setup to local Fiber nodes, then from scenarios to diagnostics, reports, route prediction, and CCH-aware comparison.

Setup and local network flow make setup installs dependencies and the fiber CLI, then fiber init and fiber up create and start local FNN nodes. Repository checkout: the starting point after git clone. repo clone make setup installs pnpm support dependencies and the local fiber binary. make setup pnpm + Cargo fiber init writes .fiber config, generated FNN configs, node keys, and Docker Compose artifacts. fiber init .fiber config fiber up starts Dockerized FNN nodes and waits for node_info RPC readiness. fiber up 3 FNN nodes fiber inspect provides a read-only view of node reachability, peer counts, and channel state. fiber inspect

Onboarding

make setup is only setup. It installs support dependencies and the local fiber CLI before any network state is created.

Scenario, diagnosis, and report flow YAML scenarios run through Fiber RPC, produce JSONL step output, attach taxonomy diagnoses, and write report artifacts. Scenario YAML describes nodes, optional channel setup, steps, expectations, and assertions. scenario YAML fiber run executes steps through FNN JSON-RPC and prints JSONL step output. fiber run JSONL steps FNN RPC returns node, channel, graph, and payment responses from the local Docker network. FNN RPC The diagnostic engine maps low-level FNN/RPC failures to taxonomy codes and remediation steps. doctor Reports write report.md, logs.json, trace.json, and last-run.json under .fiber/output. reports md/json/trace

Scenario Evidence

Every run persists its latest result. --report adds grouped causes and remediation in Markdown plus structured JSON and trace artifacts.

Route prediction and CCH comparison flow fiber predict reads local channels and graph data, scores native Fiber route confidence, and compares CCH availability without inventing live bridge data. fiber predict evaluates a configured source node, destination node, amount, and asset. fiber predict node-1 to node-2 Local channel and graph data feed the native Fiber route confidence model. route data channels + graph The analyzer returns probability, confidence label, warnings, and alternative routes when available. native score probability + fee CCH Tier 1 reports bridge availability and mechanism; live order fields require a CCH actor and Lightning backend. CCH view honest availability simulate --dry-run delegates to the same analyzer and never executes a payment. simulate --dry-run

Route Intelligence

Native Fiber gets a confidence score. CCH gets an availability/mechanism statement unless real Lightning-backed order data exists.

Watch It Work

Short terminal recordings show the core DevKit workflow without requiring judges to install Docker, Rust, or FNN locally.

Bring up a three-node Fiber network

orchestration 
Open ./casts/demo-network.cast to view the terminal recording.

Run a funded happy-path payment

scenario 
Open ./casts/demo-basic-payment.cast to view the terminal recording.

Diagnose a liquidity failure

diagnostics 
Open ./casts/demo-low-liquidity.cast to view the terminal recording.

Predict route confidence

routing 
Open ./casts/demo-predict.cast to view the terminal recording.

Explain a taxonomy code

doctor 
Open ./casts/demo-doctor.cast to view the terminal recording.

Generate reports and CI configuration

reporting 
Open ./casts/demo-report-ci.cast to view the terminal recording.

Fiber DevKit Console

Recorded replay — run fiber console locally against your own network for live data.

Loading recorded console replay.

Last Run Artifact

Try It Yourself

The fastest path is the unfunded smoke test. Funded payment scenarios require testnet CKB on generated node keys.

git clone https://github.com/anihdev/Fiber_Devkit.git
cd Fiber_Devkit
make help
make setup
fiber --help
fiber init --nodes 3 --template hub-spoke
fiber up
fiber inspect
fiber run scenarios/network-smoke.yaml --report
fiber down

Setup command

make help lists repository shortcuts. make setup installs TypeScript support dependencies with pnpm and installs the repo's Rust fiber binary locally. After crates.io publication, the install path can become cargo install fiber-devkit.

Reference Scenarios

The shipped MVP includes five reference scenarios plus one clean unfunded smoke scenario. Roadmap-only scenarios are documented separately.

Scenario Funded Keys What It Demonstrates
network-smoke.yaml No Clean generated network RPC reachability and graph sanity checks.
basic-payment.yaml Yes Funded single-hop channel setup and successful keysend payment.
multi-hop-routing.yaml Yes Hub-spoke multi-hop payment behavior across generated nodes.
low-liquidity.yaml Yes Insufficient-liquidity failure surfaced as structured diagnosis output.
peer-offline.yaml No Unreachable endpoint handling and step-level failure reporting.
channel-exhaustion.yaml Yes Low route-confidence prediction before an over-capacity payment attempt.

Category 2 Coverage

DevKit goes deep on local testing, diagnostics, route confidence, and node visibility while documenting deferred infrastructure clearly.

Strong

Local test networks, YAML scenarios, reports, CI scaffold, failure taxonomy, and fiber doctor.

Shipped Polish

fiber inspect, route prediction, graph-topology warnings, CCH comparison output, and self-hosted demo recordings.

Documented Roadmap

Live CCH order execution, alerting, full peer detail, payment-hash lookup, graph SVG, watchtower scenarios, and fee-policy simulation.

From Local Network to CI

One connected workflow carries Fiber development from repeatable local conditions to actionable results and automated integration checks.

  1. 01

    Local Network

    fiber init · up

    Start a controlled multi-node FNN environment.

  2. 02

    Scenarios

    fiber run

    Exercise successful payments and expected failures.

  3. 03

    Diagnosis

    fiber doctor

    Translate real failures into causes and remediation.

  4. 04

    Route Confidence

    fiber predict

    Assess route viability, warnings, and CCH availability.

  5. 05

    Reports

    --report

    Persist Markdown, structured logs, and trace artifacts.

  6. 06

    CI

    fiber ci init

    Carry the same repeatable checks into automation.