More screenshots
Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+22 moreClaude CodeCursorCodex CLIMuseDots
OpenClaw
Manus
Grok
Grok Bot+18 agentsCrypto Token Due Diligence — Evidence-Based Risk Check
Read-only token research with chain-specific checks, source links, and explicit unknowns. Includes an optional Python snapshot helper for Ethereum, BSC, Base, and Solana.
$9.90
Crypto Token Due Diligence — Evidence-Based Risk Check
Example session with this skill installed
Create a dated, read-only baseline report for USDC on Ethereum mainnet at 0xA0b86991c6218b36c1d19D4a2e9Eb0cE3606eB48. Verify the issuer-published address, collect selected direct contract reads and public scanner/DEX observations, and return an evidence ledger with source links, UTC timestamps and the block number. Separate observed facts from interpretations and unknowns. State which permission, sellability, liquidity and holder checks remain incomplete. Do not connect a wallet, sign, trade or claim that this baseline is a complete audit.
- Read your context and instructions
- Compiled the crypto token due
Read-only baseline: Ethereum USDC
This is the complete report from a recorded release-validation baseline. It demonstrates evidence handling; it is not a complete audit or a recommendation to buy or use USDC. A fresh run may return different data.
Target: Ethereum mainnet, chain ID 1; 0xA0b86991c6218b36c1d19D4a2e9Eb0cE3606eB48.
Asset type: Issuer-controlled stablecoin.
Snapshot: 2026-10-11T13:33:05+00:00 (UTC).
Direct state block: 26169530 (0x18f50ba), obtained as the provider's latest block. Finality and block hash were not verified; a latest block can be reorganized.
Scope: Issuer-published identity, selected bytecode/getter/storage reads, a GoPlus response and 30 DEXscreener pairs. Full implementation/role enumeration, ordinary-holder sellability, LP control and holder classification were not examined.
Assessment: INSUFFICIENT EVIDENCE for a complete before-use risk screen. Useful observations are available, but material checks remain incomplete. Missing scanner fields are not negative findings.
Coverage
| Check | Coverage | What this baseline establishes |
|---|---|---|
| Identity | COMPLETED | The queried Ethereum address matches the issuer-published target. |
| Permissions | PARTIAL | Code, selected getters and three standard slots were read. All active roles and controllers were not mapped. |
| Sellability | UNAVAILABLE | A scanner flag was returned; no relevant ordinary-holder sale or realistic swap simulation was inspected. |
| Liquidity | PARTIAL | Indexed pairs were returned; LP ownership, locks and size-specific exit depth were not verified. |
| Holders | PARTIAL | Scanner records were available; custody classification, complete exposure and a matching supply snapshot were not established. |
| Tokenomics and backing | PARTIAL | A supply getter was read. Circulating methodology, backing and issuance constraints were not independently verified. |
| Team and claims | PARTIAL | Issuer address identity was checked. Broad team, partnership and current-audit research was outside this baseline. |
Evidence ledger
CONFIRMED means the stated observation is supported by the cited source. A confirmed scanner response is not confirmation of its underlying safety claim.
| Check | Status | Observation | Interpretation and limit |
|---|---|---|---|
| Issuer identity | CONFIRMED | Circle's mainnet address table, retrieved on 2026-10-11 UTC, lists the target for Ethereum. | Establishes the issuer-published target; not reserve or safety certification. |
| Chain and block | CONFIRMED | Ethereum PublicNode RPC returned chain ID 0x1 and block 0x18f50ba at 13:33:04 UTC. | Target reads were pinned to block number 26169530, not a finalized block hash. |
| Deployed code | CONFIRMED | eth_getCode returned 2,186 bytes at 13:33:05 UTC, block 26169530. | Code exists. Verified-source correspondence and implementation behavior were not established. |
| Supply getter | CONFIRMED | At the same block/time, totalSupply() returned raw integer 48530405921259002; decimals() returned 6: 48,530,405,921.259002 units. | Exact getter observation and scaling; not independent proof of circulating supply or fiat backing. |
| Owner getter | CONFIRMED | owner() returned 0xfcb19e6a322b27c06842a71e8c725399f049ae3a at the same block/time. | A getter value alone does not identify all minters, pausers, blacklisters or upgrade controllers. |
| Standard proxy slots | CONFIRMED | The sampled ERC-1967 implementation, admin and beacon slot words were all zero at the same block/time. | This does not rule out a legacy/custom proxy or other administration. No non-upgradeability finding is made. |
| Scanner response | CONFIRMED | GoPlus token-security API reported is_proxy="1", is_honeypot="0"; owner_address and is_mintable were absent at 13:33:04 UTC. | These are observed response fields, not proof of actual sellability or a complete permission map. Scanner block unavailable. |
| Supply discrepancy | CONFIRMED | GoPlus supply was 48,528,866,271.471921, differing from the block-specific getter above. | Provider timing or caching may explain it, but that explanation remains unverified. Do not calculate holder percentages by mixing these snapshots. |
| Indexed markets | CONFIRMED | DEXscreener's token-pairs endpoint returned 30 indexed pairs at 13:33:03 UTC. | This does not prove complete venue discovery, verified LP locks or executable liquidity. Indexer block unavailable. |
| Full permissions and ordinary-holder exit | UNKNOWN | The active implementation, all material roles and relevant holder sell evidence were not examined. | Do not infer “cannot mint,” “no privileged controls” or “honeypot-free.” |
The buyer package preserves the actual requests and responses in examples/evidence/ethereum-usdc-direct.json, with the longer source report in examples/ethereum-usdc.md. Source links may return updated data; the saved JSON records this dated observation.
Findings that determine the next check
- Zero standard proxy slots do not resolve the scanner's proxy indication. Inspect the deployed proxy code, legacy/custom storage and active implementation before describing upgrade authority.
- Trace the issuer token's material minter, blacklister, pauser and upgrade controls. The owner getter is not the complete permission map.
- Establish condition-specific ordinary-holder sell evidence. A favorable scanner field does not replace an observed sale or suitable non-broadcast simulation.
- Reconcile supply observations before interpreting holder concentration.
These are research gaps. This baseline does not establish malicious behavior, full safety, reserve sufficiency or future trading access.
Next checks
Obtain deployed implementation/source correspondence and all material role/admin paths; examine relevant ordinary-holder sales or an authorized non-broadcast simulation; identify actual LP position control and usable depth for the intended size; classify holders against matching supply/time; and review backing and issuance disclosures against the active implementation.
Actions taken: Read-only public document/API/RPC queries and saved local JSON. No wallet connection, signatures, approvals, payments or transactions.
Connects securely to your tools. The creator never sees your data.
About this skill
Research a token before you commit funds, with a report you can trace back to its evidence.
Crypto Token Due Diligence guides your agent through token identity, contract permissions, holder concentration, liquidity control, tokenomics, team claims, and the gap between promotion and available evidence. It separates an observed fact from the interpretation of that fact.
What you get:
- A structured
SKILL.mdworkflow and separate EVM and Solana reference checks. - An evidence ledger recording findings as
CONFIRMED,INFERRED, orUNKNOWN, with source links and observation times. - Risk-rating rules that keep missing critical evidence visible and allow an insufficient-evidence result.
- An optional Python 3.10+ helper,
scripts/snapshot.py, using the standard library to collect a limited public-data snapshot for Ethereum, BSC, Base, or Solana. It supplements the research workflow; it does not complete every check. - Dated live examples and their stated limits.
The workflow asks who can mint, freeze, upgrade, restrict transfers, or control liquidity. It also checks whether holder percentages include exchanges, pools, or other addresses that would distort the interpretation. Wallet links, team identities, wash trading, and promotional claims remain uncertain unless the available evidence supports them.
This is read-only research. It does not connect wallets, sign transactions, buy or sell tokens, run project-supplied code, or continuously monitor a watchlist. It is not a smart-contract audit, a price forecast, or a guarantee that a token is safe.
Your agent needs browsing or search access. Python is optional. Data access depends on the host's tools and public services; blocked, rate-limited, missing, or ambiguous information is reported as UNKNOWN. Your agent provider may charge for model or tool usage.
How to use
Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scanned
Verified clean 8 days ago
- Passed all security checks, Safe to install
Requires
Dexscreener
Gopluslabs
Solana
Publicnode
Python