debug-rpc-provider
Skill76 starsChanged 20 days agoApache-2.0View on GitHub
Diagnose RPC provider health issues for a Hyperlane chain and recommend fixes. Examine the output table and classify each RPC URL: Output the following structured information: If you recommended URL changes in Step 4, end your response with the confirmation convention so the user gets an approve/reject button: This MUST be the very last thing in your message. Do NOT run /fix-rpc-urls yourself — wait for user approval. Note: [CONFIRM: ...] is a Haggis-specific harness primitive — Haggis renders it…
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 65
Remove dead/misconfigured URLs
- Distinctive: line 66
Keep healthy URLs (reorder so healthiest are first)
- Distinctive: line 68
Never leave the list empty
- Distinctive: line 84
If you recommended URL changes in Step 4, end your response with the confirmation convention so the user gets an approve/reject button:
- Distinctive: line 94
If all URLs are healthy and no changes are needed, skip this step.
- Distinctive: line 99
If ALL RPCs (private + registry) are stale/dead, the chain itself may be halted — check block explorers before recommending URL changes.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Debug RPC Provider
Diagnose RPC provider health issues for a Hyperlane chain and recommend fixes.
When to Use
- Alert-triggered: RPC error rate alerts, provider timeout alerts, stale block alerts
- User request: "debug RPC for [chain]", "why are RPCs unhealthy on [chain]"
Input Parameters
/debug-rpc-provider <chain> [environment=mainnet3]| Parameter | Required | Default | Description |
chain | Yes | - | Chain name (e.g., ethereum) |
environment | No | mainnet3 | mainnet3 or testnet4 |
Workflow
Step 1: Find Monorepo Root
MONOREPO_ROOT=$(git rev-parse --show-toplevel)Step 2: Run Debug Script
pnpm --dir "$MONOREPO_ROOT/typescript/infra" exec tsx scripts/secret-rpc-urls/debug-rpc-url-health.ts \
-e <environment> -c <chain>Step 3: Analyze Results
Examine the output table and classify each RPC URL:
| Condition | Classification |
| Health = ✅, Latency < 500ms | Healthy |
| Health = ✅, Latency >= 500ms | Slow |
| Health = ⚠️ | Stale (block age 30-120s) |
| Health = ❌, has error | Dead |
| Health = ❌, chainId mismatch | Misconfigured |
Step 4: Output Diagnosis and Recommendation
Output the following structured information:
- Summary: One-line description of the issue (e.g., "2 of 3 private RPCs are dead on ethereum")
- Current URL status table: For each private RPC URL, show status (healthy/slow/stale/dead) and key metrics
Proposed new URL list: A concrete JSON array of the recommended new private RPC URLs. This should:
- Remove dead/misconfigured URLs
- Keep healthy URLs (reorder so healthiest are first)
- Always backfill with healthy registry URLs to replace removed URLs and maintain at least the original URL count. Registry URLs are free and add redundancy.
- Never leave the list empty
Format:
Proposed new RPC URLs for <chain> (<environment>): ["https://full-url-1", "https://full-url-2"]- Suggested action: Either:
- "No action needed" if all private RPCs are healthy
- "Run
/fix-rpc-urlsto apply the proposed URL changes" if URLs need updating - "Investigate further" if the issue is unclear (e.g., all RPCs stale may indicate chain halt)
Step 5: Offer to Fix (Confirmation Gate)
If you recommended URL changes in Step 4, end your response with the confirmation convention so the user gets an approve/reject button:
[CONFIRM: Fix RPC URLs for <chain>]This MUST be the very last thing in your message. Do NOT run /fix-rpc-urls yourself — wait for user approval.
Note:
[CONFIRM: ...]is a Haggis-specific harness primitive — Haggis renders it as an inline approve/reject button. In other coding-assistant contexts it is just text, so this skill is intended to be invoked from Haggis.
If all URLs are healthy and no changes are needed, skip this step.
Important Notes
- Registry URLs are public and don't contain secrets.
- If ALL RPCs (private + registry) are stale/dead, the chain itself may be halted — check block explorers before recommending URL changes.
In this repository49 other files it gives its agents.
- AGENTS.mdAGENTS.md
- add-registry-changesetSkill
- build-docker-imageSkill
- classify-onchain-ownerSkill
- code-reviewSkill
- debug-messageSkill
- debug-validator-checkpoint-inconsistencySkill
- denylist-stuck-messagesSkill
- fetch-linear-ticketSkill
- fetch-safe-tx-batchSkill
- find-docker-imageSkill
- fix-merkle-root-mismatchSkill
- fix-rpc-urlsSkill
- gcp-logsSkill
- hyperlane-message-utilsSkill
- hyperlane-utilsSkill
- inline-pr-commentsSkill
- investigate-merkle-root-mismatchSkill
- investigate-stuck-messagesSkill
- registry-yaml-sort-policySkill
- security-reviewSkill
- self-relay-hyperlane-messageSkill
- set-moonpay-standing-quotesSkill
- start-http-registrySkill
- stop-http-registrySkill
- svm-warp-alt-manageSkill
- svm-warp-program-upgradeSkill
- tob-reviewSkill
- verify-warp-contractsSkill
- warp-balancesSkill
- warp-deploy-fund-deployerSkill
- warp-deploy-init-routeSkill
- warp-deploy-nexusSkill
- warp-deploy-register-routeSkill
- warp-deploy-select-keysSkill
- warp-deploy-send-testSkill
- warp-deploy-update-ownersSkill
- warp-deploy-validate-ownersSkill
- warp-forkSkill
- warp-key-value-expansionSkill
- warp-manage-offchain-quotesSkill
- warp-route-checkSkill
- warp-run-logSkill
- warp-simulate-svm-txsSkill
- warp-update-extendSkill
- warp-update-proposeSkill
- warp-update-resolve-artifactsSkill
- warp-updateSkill
- warp-verify-onchain-configSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

