agent-staking-mcp-server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
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": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| tool_stake_depositA | Deposit stake for an agent as a trust proof. Higher stake = higher trust tier. Tiers: unverified → bronze → silver → gold → platinum. Stake can be slashed for bad behavior. Args: agent_id: Unique identifier for the agent (e.g. DID, UUID, name) amount: Amount to stake (must be > 0) currency: Currency/token for the stake (default: REP) |
| tool_stake_verifyA | Verify an agent's stake status and trust level. Returns current balance, trust tier, slash history, and trust score. Args: agent_id: Agent identifier to look up |
| tool_reputation_by_stakeA | Get a stake-weighted reputation score for an agent (0-100). Combines stake size, slash history, and membership duration. Returns score, grade (A+ to F), and breakdown. Args: agent_id: Agent identifier to evaluate |
| tool_slash_stakeA | Slash an agent's stake for bad behavior (fraud, non-delivery, rule violations). The slash is permanently recorded in the agent's history. Slash events reduce the agent's trust score and tier. Args: agent_id: Agent to slash slash_amount: Amount to slash from their stake reason: Description of the bad behavior (logged permanently) |
| tool_dispute_openA | Open a dispute between two agents. If the defendant loses, their stake is automatically slashed by the stake_at_risk amount. Args: claimant_id: Agent filing the complaint defendant_id: Agent being accused description: What happened (will be permanently logged) stake_at_risk: Amount of defendant's stake to slash if they lose |
| tool_dispute_resolveA | Resolve an open dispute and optionally slash the loser's stake. If claimant wins, defendant's stake is automatically slashed by the disputed amount. Args: dispute_id: ID of the dispute to resolve (from dispute_open) winner_id: Agent ID of the winner (claimant or defendant) resolution_notes: Explanation of the resolution decision |
| tool_stake_leaderboardA | Get the trust leaderboard — top staked agents ranked by trust score. Shows the most trustworthy agents in the network. Higher stake + no slashes + longer membership = better ranking. Args: top_n: Number of agents to show (default: 10) |
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 7 tools
Each tool targets a distinct operation: dispute opening/resolution, stake deposit/verification/leaderboard/slashing, and reputation scoring. No overlapping or ambiguous purposes.
Most tools follow a noun_verb pattern (e.g., stake_deposit, dispute_open) or verb_noun (slash_stake). However, reputation_by_stake deviates with a preposition, and leaderboard is a noun rather than an action.
7 tools cover the core staking and dispute resolution workflows without being too few or excessive. The number feels right for the domain.
Core CRUD-like operations are present for staking (deposit, verify, slash) and disputes (open, resolve). Missing are a withdraw/unstake tool and perhaps a dispute list, but these are not essential for basic functionality.