IOTA AgentSAFE
MCP server for IOTA blockchain — 20 tools for AI agent integration (wallet, Move CLI, on-chain queries via Model Context Protocol)
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://github.com/Scottcjn/Rustchain/blob/main/BCOS.md)
MCP (Model Context Protocol) server for IOTA blockchain. Enables AI coding agents (Claude Code, Cursor, VS Code Copilot, ChatGPT) to interact directly with the IOTA network.
Tools
18 tools across 3 categories:
Wallet (8 tools)
CLI & Move (4 tools)
On-Chain Query (6 tools)
Architecture
flowchart LR client["AI coding agentClaude Code, Cursor, VS Code Copilot, ChatGPT"] mcp["iota-agent-mcp
ab70265b8156OBSERVED · 2026-10-07Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add iota-agent-mcp -- npx -y [email protected]
{
"mcpServers": {
"iota-agent-mcp": {
"command": "npx",
"args": [
"-y",
"[email protected]"
]
}
}
}Exposed tools (19)
13 read · 6 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
iota_cli | write | Run an arbitrary IOTA CLI command. Use for operations not covered by other tools. |
iota_coins | read | Get coin objects for an address (useful for gas estimation and token queries) |
iota_decompile | write | Retrieve and decompile a deployed Move module. Returns the module |
iota_epoch_info | read | Get current epoch, checkpoint, and network statistics via GraphQL |
iota_move_build | write | Build a Move package and report compilation results |
iota_move_publish_unsigned | write | Generate an unsigned publish transaction for a Move package (for agent wallet signing) |
iota_move_test_coverage | write | Run Move tests with coverage analysis. Returns test results and coverage summary. |
iota_object | read | Fetch on-chain object data by ID. Returns owner, type, version, and content. |
iota_objects_by_owner | read | List objects owned by an address |
iota_transaction | read | Fetch transaction details by digest |
iota_wallet_accounts | read | List all derived wallet accounts |
iota_wallet_address | read | Get the active wallet address |
iota_wallet_approve | read | Approve a pending signing request |
iota_wallet_balance | read | Check IOTA balance for the active wallet |
iota_wallet_pending | read | List pending signing requests awaiting approval |
iota_wallet_reject | read | Reject a pending signing request |
iota_wallet_sign_execute | write | Sign and execute a transaction via the agent wallet (human-in-the-loop approval required) |
iota_wallet_switch_network | read | Switch between mainnet, testnet, and devnet |
test_tool | read | A test tool |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- declared (3 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (1)
@modelcontextprotocol/sdk, zod, @types/node, typescript, tsx, @types/jest, jest, ts-jest
Gates applied: no_behavioural_pass.
ab70265b8156full audit observations/trust-audit/mcp-server/scottcjn__iota-agent.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-07 | ab70265b8156 | SAFE | B | 89 | first audit |
Questions
What is the IOTA Agent MCP server?
MCP server for IOTA blockchain — 20 tools for AI agent integration (wallet, Move CLI, on-chain queries via Model Context Protocol)
What tools does IOTA Agent expose?
19 in total: 13 read-only, 6 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is IOTA Agent safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does IOTA Agent need?
No credential environment variables were found in its source, so it appears to need none.
How does IOTA Agent run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as iota-agent-mcp at 1.0.0.
How current is this page?
The grade is for one exact copy of the source (ab70265b8156), read on 2026-10-07. The repository is watched and re-audited when it changes.