Rankings & Token Agents¶

Three agents for evaluating and ranking wallets and tokens using ChainAware's predictive scoring and ranking tools.

Setup required: For AI Agents — MCP registration and agent installation.

chainaware-wallet-ranker¶

Role: Ranks a list of wallet addresses by on-chain behavioural quality.

What it does: Takes a list of wallet addresses and ranks them by behavioural signals from predictive_behaviour — engagement depth, DeFi activity, tenure, transaction diversity, and activity patterns. Useful for prioritising reward allocations, governance weighting, loyalty tiers, or VIP identification across a user base.

Tools used: predictive_behaviour
Model: Claude Haiku 4.5

Example invocation:

@chainaware-wallet-ranker Rank these wallets by on-chain quality: [0xA, 0xB, 0xC, 0xD, 0xE]

Output includes:
- Ranked list of wallets (highest quality first)
- Behavioural score per wallet
- Key signals driving each rank (DeFi depth, tenure, activity)
- Tier assignments (Top 10% / Top 25% / Standard / Low engagement)
- Recommended use of the ranking (reward allocation, governance weight, etc.)

chainaware-token-ranker¶

Role: Returns and interprets a ranked list of tokens by quality and risk signals.

What it does: Calls token_rank_list to retrieve ChainAware's current token ranking data, then interprets and presents the results in a structured format. Useful for identifying the highest-quality tokens in a given universe, filtering out high-risk tokens before listing decisions, or building curated token watchlists.

Tools used: token_rank_list
Model: Claude Haiku 4.5

Example invocation:

@chainaware-token-ranker Show me the top-ranked tokens and flag any high-risk ones

Output includes:
- Ranked token list with quality scores
- Risk flags per token (rug pull risk, fraud signals)
- Top-tier tokens (recommended)
- Tokens to exclude (high risk)
- Summary statistics (distribution of quality scores)

chainaware-token-analyzer¶

Role: Deep analysis of a single token combining rank quality signals and fraud risk.

What it does: Combines token_rank_single (detailed quality ranking data for a specific token) with predictive_fraud (fraud risk on the token deployer or associated wallets) to produce a comprehensive token research report. Covers tokenomics quality signals, risk indicators, comparative rank, and an overall investment or listing recommendation.

Tools used: token_rank_single, predictive_fraud
Model: Claude Haiku 4.5

Example invocation:

@chainaware-token-analyzer Give me a full analysis of token 0xTOKEN...CONTRACT

Output includes:
- Token quality rank and score
- Rank percentile (top X% of tokens analysed)
- Quality signals (liquidity quality, holder distribution, trading patterns)
- Fraud risk on deployer and associated wallets
- Red flags (if any)
- Overall verdict: List / Watchlist / Avoid
- One-paragraph investment / listing rationale