Caelara Shield
Server Details
Screens prompts, output, URLs and EVM transactions. Heuristic. Needs an X-Caelara-Key API key.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2024-11-05
- URL
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 4 tools
Each tool targets a distinct security checkpoint: transaction/token screening, guarded HTTP fetch, prompt input scanning, and model output validation. The closest pair is scan_prompt vs validate_output, but their descriptions clearly separate input from output screening and their parameter intents are different. Minor potential confusion remains for an agent deciding whether a given piece of text is 'untrusted input' or 'model output'.
All tools use the same shield_ prefix followed by a verb_noun pattern (audit_token, proxy_fetch, scan_prompt, validate_output). Snake_case is used uniformly with no deviations or mixed conventions. The naming is predictable and readable.
Four tools for four distinct security gates is well-scoped and each tool earns its place. The count sits comfortably within a reasonable range and avoids redundancy. No tool feels like filler.
The surface covers the main agent security concerns: external fetch safety, prompt injection, token/transaction risk, and output leakage. Minor gaps exist around centralized policy management or audit logging, but the core screening lifecycle is covered and agents can work around the rest.
Available Tools
4 toolsshield_audit_tokenBInspect
Screen a token and/or an EVM transaction: token safety score, decoded transaction risks (unlimited approvals, failed or contract-creating transactions), address-poisoning lookalike check (0.050 USDC per call)
Inputs: hash, chain, token, destination, +1 more.
shield_proxy_fetchAInspect
Fetch an https URL through an SSRF-guarded proxy and return a JSON envelope with the scanned body (HTTP 200 envelope, upstream status inside) (0.025 USDC per call)
Inputs: url, allow, include_blocked.
shield_scan_promptAInspect
Screen untrusted text before it reaches a model: injection patterns, hidden characters, PII and secrets, with an allow/review/block verdict (0.015 USDC per call)
Inputs: text (required), redact.
shield_validate_outputAInspect
Check model output before it is used: secret and PII leakage, exfiltration links, active content, dangerous shell patterns, optional schema and grounding checks (0.015 USDC per call)
Inputs: output (required), schema, source, system_prompt.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Prompt injection detection API for AI agents. Scan untrusted text before passing it to an LLM.
Security scanner for AI agents: prompt injection, leaked secrets, PII, and SSRF-risk URL detection.
Scan text, documents, websites, and MCP metadata for prompt injection and sensitive-data risks.
Pay-per-call safety checks for AI agents: screen a crypto address or URL before you transact.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceInput/output safety gate for AI agents: detect prompt-injection/jailbreak, leaked secrets/PII, and URL/IP reputation. Deterministic, no LLM.1MIT
- AlicenseAqualityCmaintenanceEnables MCP-compatible coding agents to screen text for prompt injection, personal data, unsafe content, and hallucinations before acting on it, using the Altirs API.518 npmMIT

EVIDIQ Bulwarkofficial
AlicenseNot gradedqualityBmaintenanceProvides deterministic scanning and detection of prompt injection, jailbreak, data-exfiltration, and system-prompt leaks, with EIP-191 signed attestations and 0G Storage anchoring for verifiable safety reports.1MIT- AlicenseNot gradedqualityCmaintenanceEnables AI agents to run pre-sign risk checks on transactions, approvals, EIP-712 signatures, token addresses, wallet approvals, sanctions screening, and prompt-injection texts, returning green/orange/red verdicts with machine-readable reason codes before signing.1AGPL 3.0
Glama MCP Gateway
Add one secure layer between your agents and this server.