# Intake — {protocol / client name} | | | | ------------------- | ------------------------------------------------- | | **Audit #** | audit_{n} | | **Mode** | {automated (audit-cycle) / human-in-loop (audit-assist)} | | **Intake date** | {YYYY-MM-DD} | | **Answered by** | {human / defaults applied} | --- ## 1. Scope & Commit Pin | | | | ------------------- | ------------------------------------------------- | | **Repository** | {path or URL} | | **Branch** | {branch reviewed} | | **Audited commit** | `{full 40-char SHA}` (`git rev-parse HEAD`) | | **Scope** | {FULL / PROGRAM / BACKEND / FRONTEND / DEVOPS} | | **In scope** | {paths / programs / dirs reviewed} | | **Out of scope** | {excluded paths + reason — Q44} | | **audit-mem store** | {path of the store that passed the Rule 11d check / none} | --- ## 2. Target Languages & Frameworks | | | | --------------------- | ----------------------------------------------- | | **On-chain framework**| {Anchor / native Solana / Pinocchio / N/A} (Q3) | | **Languages detected**| {Rust / TypeScript / Python / Go / …} | | **Backend** | {Express / Fastify / NestJS / none / …} (Q4) | | **Frontend** | {Next.js / React / Vue / none / …} (Q5) | | **Database** | {PostgreSQL / MongoDB / none / …} (Q6) | | **Monorepo** | {yes / no / multi-repo} (Q7) | --- ## 3. Protocol Class | | | | ----------------------- | --------------------------------------------- | | **Project type** | {DeFi / NFT / DAO / payments / gaming / infra / …} (Q1) | | **Handles user funds** | {yes — custodial / yes — non-custodial / no} (Q17) | | **Tokens supported** | {SOL / specific SPL / any SPL / Token-2022 / …} (Q18) | | **DEX integrations** | {Jupiter / Raydium / Orca / none / …} (Q19) | | **Fees** | {management / performance / swap / withdrawal / none} (Q20) | | **Withdrawal model** | {instant / multi-step / time-locked / admin-approved / none} (Q21) | | **Uses oracles** | {Pyth / Switchboard / custom / DEX-quote / none} (Q23) | | **CPI targets** | {Token / ATA / DEX / staking / other / none} (Q24) | | **Economic review** | {required (funds=yes) / not required} — checklist 06 {on/off} | --- ## 4. Compliance Frameworks | | | | ----------------------- | --------------------------------------------- | | **Regulatory scope** | {none / MiCA / SEC / GDPR / SOC 2 / DORA / …} (Q35) | | **Collects PII** | {no — wallets only / email / KYC / financial} (Q36) | | **Compliance checklist**| {checklist 18 on / off} | --- ## 5. Severity Calibration | | | | ------------------------- | ------------------------------------------- | | **Deployment status** | {pre-launch / devnet / mainnet-live} (Q8) | | **TVL / funds at risk** | {$0 / <$10K / $10K-100K / $100K-1M / $1M-10M / >$10M} (Q10) | | **Upgradeable** | {single wallet / multisig / DAO / immutable / N/A} (Q11) | | **Top concerns (ranked)** | {1. … / 2. … / 3. …} (Q40) | | **Audit depth** | {quick / standard / deep / maximum} (Q41) | **Calibration rules applied** - {Q8 = mainnet-live → +1 severity on all fund-related findings.} - {Q10 > $1M → critical findings double-weighted in risk score.} - {Q11 = single wallet → upgrade authority auto-flagged Severity 8+.} - {Q41 = quick → grep + known-vectors only, semantic analysis skipped (state as coverage limit).} --- ## 6. Trust-Model Inputs (CRITICAL) | Actor | Who / how gated | Trusted to (privileges) | Trusted NOT to | | ----- | --------------- | ----------------------- | -------------- | | **Upgrade authority** | {single wallet / multisig / DAO / immutable} (Q11) | {upgrade program} | {push a malicious upgrade; assumed key-secure} | | **Admin / manager** | {wallet sig / admin JWT / on-chain authority} (Q16) | {set params, pause, …} | {abuse privileged ops; assumed honest} | | **Oracle** | {Pyth / Switchboard / custom} (Q23) | {supply price} | {report manipulated / stale price beyond bounds} | | **Keeper / crank** | {who runs it} | {trigger scheduled ops} | {reorder / withhold to grief} | | **LP / depositor** | {permissionless} | {deposit, withdraw own} | — (untrusted) | | **Permissionless user** | {anyone} | {call open instructions} | — (untrusted — the attacker) | **Trust-model narrative** — {2-4 sentences: the security of the system depends on which of the above holding. State the load-bearing ones plainly. This is the seed text for report §4.6.} --- ## 7. Security History | | | | ----------------------- | --------------------------------------------- | | **Prior audits** | {firm name / internal / tools only / first audit} (Q25) | | **Incidents** | {fund loss / blocked exploit / breach / none} (Q26) | | **Bug bounty** | {public / private / none} (Q27) | --- ## 8. Execution Mode & Outside Services | | | | ------------------------- | ------------------------------------------- | | **Execution mode** | {isolated (recommended) / native} (Q46) | | **Isolation used** | {container / VM / Claude Code sandbox — no keys or credentials mounted / N/A (native)} | | **Outside services** | {disclosed in the report per Rule 11b — e.g. RPC provider, second-opinion LLM, registry queries / none planned} | --- ## 9. Assumed Defaults (Unanswered) | Field (Q#) | Default applied | Basis | | ---------- | --------------- | ----- | | {Q10 TVL — unanswered} | {assumed "Unknown" → severity calibrated to pre-launch baseline} | QUESTIONS.md default | | {Q11 upgradeable — unanswered} | {assumed present; upgrade authority treated as trusted-but-single unless code shows otherwise} | QUESTIONS.md default | | {Q23 oracle — unanswered} | {assumed no oracle unless code shows a feed} | inferred from code | | {Q46 execution mode — unanswered} | {native — target-controlled code runs on the auditor's machine; the report states it} | OUTPUT-RULES Rule 11a default | | {…} | {…} | {…} | > Every default listed here is carried into the report's **§4.6 Assumptions & > Simplifications** as an explicit assumption — so "no finding here" reads against > the stated default rather than as a blanket clearance.