web3-pipeline

Guides development of an 11-chain Proof-of-Growth Web3 pipeline with Solidity contracts and Rails services.

23|1|Updated Feb 26, 2026
One-click install
npx skills add https://github.com/Alexey-Lukin/silken_net --skill web3-pipeline-alexey-lukin
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: web3-pipeline
Source: https://github.com/Alexey-Lukin/silken_net/tree/main/.claude/skills/web3-pipeline
Command: npx skills add https://github.com/Alexey-Lukin/silken_net --skill web3-pipeline-alexey-lukin

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Working on the silken_net Web3 surface — minting, slashing, payouts, the Ethereum L1 anchor, and DAO governance — involves dozens of non-obvious invariants (double-spend guards, unit/direction traps, activation-gated oracle legs) where a wrong assumption silently moves real value. This Skill routes you to the correct canonical docs and surfaces the gotchas before you touch any money path. ## Core Features & Use Cases - Canonical doc routing: Maps every Web3 task to its single source of truth across CLAUDE.md sections and the 05_01–05_06 documentation canon (multichain architecture, Proof-of-Growth pipeline, SCC/SFC tokenomics, L1 anchoring, slashing policy, DAO governance). - Indexed gotcha knowledge base: Ships a 38+ item append-only gotcha index (bodies in gotchas.md) covering WEB3_STRICT_MODE semantics, manual_review double-spend guards, batchMint bisection, Dynamic Tax rules, Solana Ed25519 payouts, and slashing penalty-factor de-correlation. - Common task playbooks: Step-by-step guidance for adding a chain integration, changing minting thresholds, and editing Solidity contracts under the Foundry/Slither/Halmos/Medusa audit gates with gas-snapshot baseline rules. - Use Case: You need to change the minting threshold or debug a transaction stuck in manual_review — the Skill tells you exactly which doc owns the mechanic, which guard clauses apply, and which traps (e.g., beneficiary-side KYC gating, gross vs available balance) will bite. ## Quick Start Ask the assistant to help you add a new chain integration or explain why a blockchain transaction is stuck in manual_review on the silken_net Web3 pipeline.

Frequently Asked Questions about web3-pipeline

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I add a new blockchain integration to a Rails Web3 backend?

Create a service in app/services/, add a worker, configure ENV variables, and verify the guard clauses, then update the multichain architecture doc and the routing table. The Skill maps each step to its canonical documentation home.

Why is a blockchain transaction stuck in manual_review?

manual_review is a double-spend guard, not a review queue: a tx_hash exists but the on-chain outcome is unknown, so funds stay locked until manual verification. It is age-unbounded by design, and insurance reserve holds also park rows there.

How does batchMint handle a reverting batch?

The minting service dry-runs first, then binary-searches the reverting batch (max depth 6) to isolate the poisoned record, bisecting only within one archive-root subgroup. Never bypass the dry-run step.

Does WEB3_STRICT_MODE control production safety behavior?

No — it is belt-and-suspenders, not the switch. The Hadron stub, callback HMAC, and IoTeX fallback all fail closed in production regardless of the flag, and a separate WEB3_CHAIN_ENV variable now scopes mainnet versus testnet assertions.

What signing algorithm do Solana micro-rewards use?

Solana rewards use Ed25519 signing, not secp256k1, with the associated token account resolved by owner. The reward formula is 10,000 plus growth_points times 100 lamports, flipping to hourly batch payouts when the USDC threshold parameter exceeds zero.

When does slashing fire on the Proof-of-Growth pipeline?

Slashing fires only on positive Category-A evidence — direct tamper proof via the positive_a? predicate. Indirect signals route to frozen status plus a field audit instead, and correlated uplift signals combine with max(), never a sum.