All servers / Other / Caelara Shield
Caelara Shield
Screens prompts, output, URLs, EVM transactions and swaps. Heuristic. Free key at caelaralabs.com.
Tested by ConnectATD Hosted serverTested 2026-11-22. Address: https://shield.caelaralabs.com/mcp · Registry name: com.caelaralabs/shield
Score 95 out of 100. Tested 2026-11-22.
Score breakdown
| Connects and stays up | 20 / 20 | |
|---|---|---|
| Tool descriptions | 20 / 20 | |
| Inputs explained | 15 / 15 | |
| Speed | 15 / 15 | |
| Safety basics | 10 / 15 | |
| Upkeep | 15 / 15 |
What we saw
- Connects and answers the MCP handshake: yes (0.5 seconds)
- Uses an encrypted (https) connection: yes
- Server reports itself as: Caelara Shield 1.0.0
- Tools offered: 7 · Parameters with a plain description: 36 of 37
- Official registry entry last updated: 2026-11-20
From the publisher
- No notes.
How hosted servers are graded
Hosted servers run on the publisher's own machines, so we cannot score downloads or install safety. We use six checks instead (connects and stays up 20, tool descriptions 20, inputs explained 15, speed 15, safety basics 15, upkeep 15). Speed is the first call to a safe, read-only tool (when a server marks none, the time to connect and list its tools), using a fresh request where we can so a saved answer cannot hide a slow one. We never call tools that write data or cost money. A server that needs a login or key is not graded. Full method.
Want this server set up for your team?
ConnectATD builds AI automation and wires up MCP servers like this one. Tell us what you are trying to do.
Talk to ConnectATDTools and how they are described
Agents choose tools from these descriptions.
shield_scan_prompt2 of 2 parameters describedScreen 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)
shield_validate_output4 of 4 parameters describedCheck 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)
shield_proxy_fetch3 of 3 parameters describedFetch 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)
shield_audit_token5 of 5 parameters describedScreen 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)
shield_route_quote10 of 10 parameters describedScreen and quote a token route (same-chain or cross-chain) across Solana, Ethereum, Base, Arbitrum, Optimism, Polygon and BNB, racing several providers with a 50 bps fee. Returns risk telemetry; set strictBlock true to refuse high-risk routes. Free. Heuristic screening — will miss attacks.
shield_safe_swap10 of 10 parameters describedScreen a token route, then build unsigned transactions for the best route (needs taker; recipient for EVM <-> Solana). The caller signs. Set strictBlock true to refuse high-risk routes. Free. Heuristic screening — will miss attacks.
shield_universal_action2 of 3 parameters describedPre-execution guardrail for any wallet action: screens it, then builds an unsigned payload the caller signs. type: token_route (multi-chain token routes, 50 bps), perp_order (Hyperliquid builder code or GMX v2 UI fee), prediction_order (Polymarket CLOB order), nft_action (Zora mint with mint referral, or a Seaport listing check), tx_guardrail (decode arbitrary calldata: approvals, operators, transfers, unknown calls). strictBlock true refuses riskLevel high with 422 policy_blocked. Free. Heuristic screening — will miss attacks.
Tested by ConnectATD. Spotted a mistake? Tell us.