Sandbox-truth probe for agent frameworks
Server Details
Measured sandbox verdicts: 7-axis probe, signed pinned receipt. Cached $0.25, fresh $2.50.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-03-26
- URL
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 5 tools
The tools mostly target distinct operations: framework listing, gate evaluation, receipt verification, cached verdict lookup, and watch status. The main overlap is between sandbox_cached_verdict and sandbox_watch_status, since both surface cached verdict information, but their descriptions make the distinction reasonably clear.
All tools share the sandbox_ prefix and use snake_case, giving a predictable namespace. The suffix conventions are slightly mixed—some are noun-like (frameworks, gate, watch_status) while one is verb_noun (verify_receipt)—but the set remains readable and mostly consistent.
Five tools is well-scoped for a focused sandbox truth/probe server. Each tool appears to earn its place by covering discovery, evaluation, verification, lookup, and watch state without obvious padding.
The surface covers the core probe workflow: listing fixtures, evaluating gate conditions, verifying receipts, checking cached verdicts, and viewing watch status. Minor gaps may exist around actively submitting or refreshing a probe/watch, but the essential verification lifecycle is represented.
Available Tools
5 toolssandbox_cached_verdictCInspect
Look up whether a cached verdict exists (delivery is paid over x402 on the HTTP surface).
Inputs: framework (required).
sandbox_frameworksCInspect
List probeable fixtures and the real-population status.
Inputs: no named parameters.
sandbox_gateCInspect
Dispatch-gate evaluation: pin, DEAD-FLAG, stale-hash, freshness and framework refusal (fail-closed allowlist).
Inputs: receipt (required), max_age_seconds, expected_framework, expected_source_hash.
sandbox_verify_receiptCInspect
Verify a receipt against the pinned signer key.
Inputs: receipt (required).
sandbox_watch_statusCInspect
Your watch subscriptions and current cached verdicts.
Inputs: no named parameters.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Dependency-probe-gated exit execution with hash-verified extraction and signed receipts.
Permission before writes. Gravity→DENYs→Session $0.25. Pilot $100; Gateway $299. Probe free.
Signed telemetry origin verdicts, intake census, canary drills. Verdict $0.01/event.
Live-payload conformance probes with signed receipts: ABSENT/EMPTY/PARTIAL/COMPLETE/BROKEN.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceEd25519-signed market-state receipts for 28 exchanges: is the venue open right now, as a receipt any agent can verify without trusting the operator, fail-closed (unknown is reported as closed). MCP tools: get_market_status, get_market_schedule, list_exchanges, get_payment_options. Free tier 500 calls a day with an instant key; x402 pay-per-call at 0.001 USDC on Base; Builder 99 USDC a monthMIT
- AlicenseBqualityBmaintenanceThis server enables decidable, hardware-attested semantic verification of AI outputs using on-chip ballistic walks, providing reproducible receipts that can be recomputed byte-for-byte.12345 npm1MIT
- FlicenseNot gradedqualityDmaintenanceCryptographically anchored, tamper-evident evidence receipts for AI agents — verified run receipts, existence-at-time proofs, and cited answers from an anchored public record. Remote MCP with proof-gated settlement; attests existence and integrity, never truth.-

CodexRun Ledger MCPofficial
FlicenseNot gradedqualityDmaintenancePaid remote MCP for generating trustworthy code run receipts, scope summaries, missing evidence checks, and client handoff exports.-
Glama MCP Gateway
Add one secure layer between your agents and this server.