Insight Pre-Trade Safety
Server Details
Decision-input integrity for onchain AI agents: verify oracle prices before signing a trade.
The last two health checks of this connector failed. If this server requires authentication, missing or expired test credentials may be the cause. A test profile lets Glama authenticate for health checks and discover tools; it is separate from your personal connections.
If you are the author, claim ownership, then add or update a test profile under Admin → Test Profile.
- Status
- Unhealthy
- Uptime
- 0.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- Repository
- imokokok/Insight
- GitHub Stars
- 7
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 4 tools
The four tools map to distinct lifecycle stages: monitoring, pre-trade guard/preparation, position lifecycle, and post-trade evidence. However, there is some overlap between trade_guard's check_safety/prepare ops and position_lifecycle's guard/receipt steps, and between oracle_watch_window's poll/close and position_lifecycle's close. The descriptions explicitly warn against these misuses, which keeps the boundary workable but not perfectly crisp.
All tool names use consistent snake_case naming and are descriptive noun phrases tied to their domain role. Even though they are not verb_noun action names, the convention is uniform across the set.
Four tools is a reasonable count for the lifecycle-oriented purpose and each tool clearly occupies a needed stage. The trade_guard tool does absorb a very large number of operations and deprecated aliases, making its effective surface much broader than the raw count suggests.
The surface covers the full pre-trade lifecycle: trust monitoring, oracle/market reads, safety checks, transaction preparation, position lifecycle management, and post-trade evidence verification. No obvious operational dead ends are apparent for the stated domain.
Available Tools
4 toolsoracle_watch_windowAInspect
Oracle Watch for monitored trust signals. Use it to read one trust snapshot or history series, buy a 24h monitored window, or poll/close an already purchased window. Each call selects the branch with "op": snapshot, history, begin, poll or close. Do not use it for pre-trade safety (use trade_guard op check_safety), do not buy a second window for the same subject, and never reuse a token after close. Returns the trust signal or history series, the windowId plus one-time oww_ token and expiry for begin, and the poll signal or combined close summary for window ops. Absorbs the deprecated names oracle_watch, oracle_watch_window_begin and oracle_watch_window_control. Metering: begin is 500 credits / $0.500 for the 24h window; in-window poll and close are covered (0 credits); snapshot and history are C3 (20 credits / $0.020).
Inputs: op (required), days, chain, symbol, +3 more.
position_lifecycleAInspect
Use it to buy a 24h Position Lifecycle bundle and run the guard, receipt and close steps inside it. Each call selects the branch with "op": begin, guard, receipt or close. Do not use it for a single pre-trade guard (use trade_guard op check_safety), for unrelated positions, or with a token after close. Returns the lifecycleId plus one-time plc_ token and expiry for begin, and the guard verdict, signed execution receipt or combined lifecycle summary for the in-window steps. Absorbs the deprecated names position_lifecycle_begin and position_lifecycle_step.
Inputs: op (required), chain, family, params, +5 more.
trade_guardAInspect
Use it for every pre-execution trust decision: one oracle price, price analytics, oracle health, oracle registry search, a venue market quote, or the pre-trade safety gate including prepare_swap and prepare_perp_order. Each call selects the branch with "op". Do not use it to prove execution after the fact (use trade_proof), to poll a prepaid monitoring window (use oracle_watch_window), or to manage a position bundle (use position_lifecycle). Returns the op payload: price and analytics rows for read ops, a venue quote for market_quote, a PASS/CAUTION/DANGER/BLOCK verdict with reasons for check_safety, and an unsigned prepared transaction, Solana message, venue order or instruction for prepare_* ops. Absorbs the deprecated names get_oracle_price, get_price_analytics, get_oracle_health, search_oracle_registry, get_market_quote, check_trade_safety, prepare_swap and prepare_perp_order. Metering: oracle_price, market_quote and search_registry are 2 credits ($0.002); price_analytics and oracle_health are 4 credits ($0.004); check_safety, prepare_swap and prepare_perp_order are 40 credits ($0.040), which is both the per-call price and the product ceiling. Deprecated alias spellings are callable at the same price until 2027-01-06.
Inputs: op (required), to, date, days, +69 more.
trade_proofAInspect
Use it to open a trade-evidence session, issue an execution receipt, or verify an evidence artifact after a trade. Each call selects the branch with "op": begin_evidence, issue_receipt or verify_evidence. Do not use it to decide whether a trade is safe before signing (use trade_guard) and do not use it to settle a swap guard (use the venue guard tools). Returns the evidence session handle, the signed receipt, or the verification verdict, digest and matched evidence. Absorbs the deprecated names begin_trade_evidence, issue_execution_receipt and verify_evidence. Metering: begin_evidence and issue_receipt are C4 (40 credits / $0.040); verify_evidence is free for the three offline kinds (execution_pair, base_agent_swap, solana_swap) and C4 for live reads.
Inputs: op (required), kind, mode, user, +21 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Signed Bitcoin and crypto prices for AI agents: multi-exchange prices anyone can verify offline.
Agent-native price and event oracle with cryptographic source attestation
Safety checks for AI agents: preflight before signing, prompt-injection scan, sanctions, rug checks.
Axiom Oracle — Solana-native attestation infrastructure for AI agents. RWA data on-chain.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceVerifiable action receipts for AI agents — agents sign claims locally, an independent witness countersigns and timestamps, anyone can verify offline.9 npmMIT

evermint-mcpofficial
AlicenseNot gradedqualityDmaintenanceTamper-evident receipts for AI agent actions. The notary layer for agent-to-agent transactions.1MIT- AlicenseNot gradedqualityDmaintenanceProvides cryptographic governance receipts for AI agents, enabling pre-execution evaluation and signed verdicts (EXECUTE/BLOCK/REVIEW/SHADOW) with offline-verifiable audit trails.MIT
- AlicenseAqualityBmaintenanceReal-time on-chain alpha, anti-honeypot contract security audits, and prediction market arbitrage signals for autonomous AI agents.65124 npmMIT
Glama MCP Gateway
Add one secure layer between your agents and this server.