djd-agent-score-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| PORT | No | HTTP server port (only used when TRANSPORT=http) | 3000 |
| TRANSPORT | No | Transport mode: stdio or http | stdio |
| DJD_BASE_URL | No | API base URL (use http://localhost:3001 for local dev) | https://djd-agent-score.fly.dev |
| DJD_TIMEOUT_MS | No | Request timeout in milliseconds | 10000 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| score_basicA | Get the basic reputation score for an AI agent wallet on Base. Returns a numeric score (0-1000), tier (e.g. "Trusted", "Neutral", "Risky"), confidence level, recommendation text, model version, and freshness info. This is a FREE endpoint — no x402 payment required. Args:
Returns: { wallet, score, tier, confidence, recommendation, modelVersion, lastUpdated, computedAt, scoreFreshness } Examples:
|
| score_fullA | Get the full reputation score with dimension breakdown for an AI agent wallet. Returns everything from basic score PLUS per-dimension scores and rich data:
The identity dimension includes "insumerVerified" (v2.3+) which indicates whether the wallet holds qualifying tokens verified via the Insumer attestation API, contributing up to 15 points toward the identity score. PAID endpoint — requires x402 payment ($0.10 USD). If your agent framework supports x402, the 402 response will contain payment instructions. Complete the payment and retry the request. Args:
Returns: { wallet, score, tier, confidence, recommendation, modelVersion, lastUpdated, computedAt, scoreFreshness, sybilFlag, gamingIndicators, dimensions, dataAvailability, improvementPath, scoreHistory } |
| score_refreshA | Force a re-score of an AI agent wallet using the latest on-chain data. Use this when you suspect the cached score is stale or after known on-chain activity that should change the score. PAID endpoint — requires x402 payment ($0.25 USD). Args:
Returns: { score, tier, confidence, recommendation, modelVersion, refreshedAt } |
| report_fraudA | Submit a fraud report for a wallet with supporting transaction hashes and evidence. PAID endpoint — requires x402 payment ($0.02 USD). Args:
Returns: { success, reportId, message } |
| check_blacklistA | Check if a wallet has any fraud reports filed against it. PAID endpoint — requires x402 payment ($0.05 USD). Args:
Returns: { wallet, reported, reportCount, reports[] } |
| get_badgeA | Get the embeddable SVG badge URL for a wallet's reputation score. Returns the badge endpoint URL and the raw SVG content. The URL can be embedded in markdown, HTML, or any context that supports images. This is a FREE endpoint. Args:
Returns: { badgeUrl, svg } |
| get_leaderboardA | Get the leaderboard of top-scored AI agent wallets. Returns a ranked list of wallets with their scores and tiers. This is a FREE endpoint. Args:
Returns: Array of { wallet, score, tier } |
| register_agentA | Register an AI agent wallet with metadata (name, description, optional GitHub URL). This is a FREE endpoint. Args:
Returns: { success, message } |
| health_checkA | Check the DJD Agent Score API system status. This is a FREE endpoint. Returns: { status, version, uptime } |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 9 tools
Each tool has a distinct, non-overlapping purpose. The three score variants (basic, full, refresh) are clearly differentiated by granularity and intent, while check_blacklist, report_fraud, get_badge, and register_agent target completely different resources with no ambiguity.
Most tools follow a clear verb_noun pattern (check_blacklist, get_badge, register_agent) or a consistent domain prefix pattern (score_basic, score_full, score_refresh). The only deviation is health_check, which reverses the pattern used by check_blacklist and should ideally be check_health.
Nine tools is an ideal scope for an agent reputation service, covering the full lifecycle: registration, scoring (basic/full/refresh), fraud reporting/checking, discovery (leaderboard), visualization (badge), and system status. No tool feels redundant or superfluous.
The surface covers core workflows well: CRUD-like operations for scores and fraud reports, registration, and discovery. Minor gaps exist: no update_agent to modify registered metadata after creation, and no delete/deregister capability, though agents can work around these limitations.