Audit ContestsAvailability: Intake open
Find the parts of your code that previous audits missed.
Focus the next review where it matters. A gap review maps your last audit onto the code that is live now. A full-code contest reviews frozen code at four depths, each with its own owner.
Remaining gap · Pendle V2 · RouterV4 + 7 facets
File that no report we hold names
lib/pendle-core-v2/contracts/router/ActionMiscV3.sol
20callable functions
58 of 58 callable functions on Pendle have no report naming their file.
Reports checked for Pendle
- ChainSecurity, Pendle V2 code assessment63e08e6937…
Functions in this file
- boostMarkets(address[])
- burnAccount()
- callAndReflect(address,bytes,bytes,bytes)
- exitPostExpToSy(address,address,uint256,uint256,uint256)
- exitPostExpToToken(address,address,uint256,uint256,(address,uint256,address,address,(uint8,address,bytes,bool)))
- exitPreExpToSy(address,address,uint256,uint256,uint256,uint256,(address,uint256,((uint256,uint256,uint256,uint8,address,address,address,address,uint256,uint256,uint256,bytes),bytes,uint256)[],((uint256,uint256,uint256,uint8,address,address,address,address,uint256,uint256,uint256,bytes),bytes,uint256)[],bytes))
- exitPreExpToToken(address,address,uint256,uint256,uint256,(address,uint256,address,address,(uint8,address,bytes,bool)),(address,uint256,((uint256,uint256,uint256,uint8,address,address,address,address,uint256,uint256,uint256,bytes),bytes,uint256)[],((uint256,uint256,uint256,uint8,address,address,address,address,uint256,uint256,uint256,bytes),bytes,uint256)[],bytes))
- mintPyFromSy(address,address,uint256,uint256)
- mintPyFromToken(address,address,uint256,(address,uint256,address,address,(uint8,address,bytes,bool)))
- mintSyFromToken(address,address,uint256,(address,uint256,address,address,(uint8,address,bytes,bool)))
Remaining gaps on the other protocols
| Protocol | File with the longest function list | Gap files | Functions |
|---|---|---|---|
| EigenLayer | TransparentUpgradeableProxy.sol+ 2 more files | 10 | |
| Uniswap | Owned.sol | 2 | |
| Aave | InitializableUpgradeabilityProxy.sol+ 1 more files | 2 |
In audit scopeNo gap left: every callable function of Compound, Morpho, Lido, Safe, Sky and Euler sits in a file that a report we hold names.
Recomputed with every cited report we hold, not only the first one collected. A gap means no report in our evidence names the file. It does not mean the code was never audited, and it is not a finding.
In audit scope means a report names the file. The deployed code has not been compared with the reviewed code.
- Recorded data
- 656callable functions
- 584in audit scope
- 72no audit found
- 0not checked
- 20cited reports
02Two review types
Gap review and full-code contest, side by side.
Both start from your repository. They differ in what is reviewed and what you get back.
| Question | Gap review for deployed code | Full-code audit contest |
|---|---|---|
| What is reviewed | Gap review for deployed codeThe current deployment compared with the latest relevant audit: new contracts, changed code, proxies and who can change them. | Full-code audit contestA frozen commit. Whole functions, whole contracts, the paths between contracts, and the system as a whole. |
| What you bring | Gap review for deployed codeYour deployed addresses and the audit reports you already have. | Full-code audit contestThe repository, the commit to freeze and the scope you want reviewed. |
| Who reviews | Gap review for deployed codeProof of Audits maps the coverage. Any gap you choose to close goes to a confirmed gap audit. | Full-code audit contestApproved auditors matched by tier, skills, conflicts and availability, with independent validation up the chain. |
| What you get | Gap review for deployed codeA map of what the last audit covers in production and what still needs review. | Full-code audit contestA locked report tied to the exact reviewed commit, then a fix check after you patch. |
03Full-code contest
How a full-code contest runs.
Three phases. The middle one is the contest itself, split into four tiers that each own a different depth of the code.
01Prepare
Pre-audit
Pre-audit prepares the contest: freeze the code, map the risks, and write a brief so reviewers do not start cold.
02Review
Core audit
Core audit is the contest. Function, contract, connections, and the whole system each have an owner. Findings get checked up the chain. The locked report feeds the live page.
Core audit tiers
- T4
- Function behaviorEntry functions, local invariants, asset movement, access controls, and owned helper closures.
- T3
- Whole-contract reviewState assumptions, permissions, external calls, and cross-function behavior without fragment splits.
- T2
- Cross-contract pathsRelated contract pairs, dependencies, or integrations with evidence-backed relation gates.
- T1
- Whole-system reviewSystem economics, governance, emergent risk, and the one full-system cluster.
Findings move up the chain: T4 to T3, T3 to T2, T2 to T1, then a final review closes the report and any dispute.
03Verify
Post-audit
After the team patches the bugs, we replay them on the new code and say yes or no. A fix is not done until that check is clear.
04Researchers
For researchers.
You review whole functions and whole contracts, never line fragments. Your tier decides which depth you can own.
- Eligibility
- Apply once. Assignments check your tier, relevant skills, conflicts of interest and availability for the exact scope.
- Scope
- Each assignment is a whole entry function and its helpers, a whole contract, a set of related contracts, or the full system.
- Findings
- Each submission is validated by the tier above it, then a final review decides the report and any dispute.
05Listings
Contests
Contest listings are not public yet.
Not sure yet?
List your protocol first. Users then see whether audit evidence exists for your contracts, and you can request a review later.