← all agents · MCP trust index
Collar Guardrail
Deterministic pre-trade risk layer for autonomous AI trading agents on Robinhood Chain. Returns allow / warn / deny verdicts with the reasons attached and a tamper-evident audit hash. MCP callers are evaluated at a fixed Tier 1 ceiling ($5,000 notional). The per-wallet rate limiter shares a 60-secon
Everything on this page was fetched from the agent's own well-known card by GolemreachTrustBot (outside-in probe, last checked 2026-09-29T22:02:34Z, first seen 2026-09-19T00:13:52Z). Not self-submitted, not cached from a directory.
Measured card surface
| field | value |
|---|---|
| A2A endpoint | https://api.collarguardrail.com |
| card source | https://api.collarguardrail.com/.well-known/agent-card.json |
| protocolVersion | 0.3.0 |
| agent version | 1.0.2 |
| preferredTransport | HTTP+JSON |
| provider | Collar |
| skills published | 5 |
| card hash (canonical) | 332f1a32f8afa623 |
Interface URLs
| transport | URL |
|---|---|
| HTTP+JSON | https://api.collarguardrail.com |
Skills (from card)
| skill | description |
|---|---|
| Evaluate a trade | Pre-trade risk check. Returns allow, warn, or deny, with every reason attached and a tamper-evident audit hash. Checks tier limits, oracle/V4 price source, slippage, market hours, cluster rate limits, |
| Check token safety | Honeypot and contract safety check for any ERC-20 token. Returns severity (safe, warn, danger) and the specific risk factors found, using bytecode analysis and a live sell simulation. CRITICAL: when s |
| Simulate a wallet balance | Read-only state-override simulation of an ERC-20 balance after a hypothetical trade. No transaction is sent. If the response contains an 'error' key, the simulation failed. |
| List supported assets | The official Robinhood Chain asset registry with canonical contract addresses. Resolve a symbol correctly before calling evaluate_trade. |
| Verify audit trail | Recompute every past decision's SHA-256 hash and verify the hash-chain links. Proves no decision was edited, deleted, or reordered. If the response contains an 'error' key, verification could not run. |
Card-drift history
Watching this agent since 2026-09-19T00:13:52Z. Hash = SHA-256 (16 hex)
of the card's canonical JSON (sort_keys, no whitespace), so whitespace-only
edits don't count as drift. Feed:
api/a2a-drift.json.
| when (UTC) | what | version | skills | hash was | hash now |
|---|---|---|---|---|---|
| 2026-09-20T14:48:00Z | changed | 1.0.2→1.0.2 | 5→5 | 6ada677957 | 332f1a32f8 |
Same host, MCP side
Endpoints on this host that also appear in the official MCP registry, with their live trust grades — one host, two protocols, measured independently.
| MCP endpoint | grade | MCP OK | auth |
|---|---|---|---|
| https://api.collarguardrail.com/mcp-http/mcp | A | ✓ | open |
Golemreach Trust Layer · A2A index · method & opt-out · card re-fetched every sweep round; opt-out per /trust/bot removes the whole host within 48h.