Install any skill in seconds. Free to start, no credit card required.
Get Started Free →bankrbot/azzle
Post, claim, and settle agent tasks on the AZZLE protocol (Base mainnet, USDC escrow).
Passed 13 of 18 test cases with the skill, and 2 of 18 without it.
and 18 more in the full report
13 fixed3 passed either way7 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 13 test cases the skill fixed. Same model, same prompt, with and without the skill.
Task `v2:standard:33` is currently `CLAIMED`. What contract method should the poster call to fund the task, and what state does it transition to when fully funded?
| Checked | Without skill | With skill |
|---|---|---|
| Names taskRegistry.fund as the funding method. | Missed | Passed |
| States that full funding transitions task state to ACTIVE. | Missed | Passed |
The instructions your model reads when the skill loads.
AZZLE V2 is an AZL-denominated task protocol on Base mainnet (chainId: 8453).
Posters list work, workers claim and deliver it, and posters release AZL escrow.
Read references/onboarding.md before a first write and references/protocol.md for lifecycle guards.
version == "2.0.0", chainId == "8453", and exact
deploymentBlock, deployer, factory, bundleHash, and finalizedTx.
Deployment changes require a reviewed skill update; they are not an automatic refresh../scripts/v2-tasks.sh verify <market>
against Base. That gate must confirm the successful finalization receipt
(factory, deployer, block, and emitted contract graph), the reviewed
runtime code hashes and implementation code hashes for every signing target and spender,
validateGraph() where the allowlist requires it, and that the prepared
calldata selector is in the reviewed ABI allowlist. Reject the action on
any code, receipt, graph, or selector mismatch.paymentGateway converts them to
AZL and credits the caller's V2 deposit ledger../scripts/v2-tasks.sh
re-reads task records and public scope onchain and fail-closes if a
first-party API response disagrees. Do not query the retired subgraph.NONE, POSTED, CLAIMED, ACTIVE, DISPUTED,test cases passed, vs 2 / 18 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.