Loading…
Runs verified read-only code tools without source exposure: backtest stats, strategy gates, scans.
Do not connect
The assessed surface is high-risk. Remediate the findings before connecting.
Scanned 1 month ago Due for re-check
A server can change after it's graded. Re-run the automated scan to refresh this report.
This grade is deterministic and reproducible: the same server surface always yields the same grade under a given algorithm version. It is a real automated assessment computed by the MCPGrade engine from what the probe actually observed — not a fabricated or opinion score. It is not a manual human pentest, so it can miss context-specific risks.
Every signal below was measured directly by the automated probe. The grade is derived only from evidence like this — nothing is assumed.
The Apex (smartapex.uk) MCP server exposes 24 tools, focused primarily on developer capabilities. Its published description reads: "Runs verified read-only code tools without source exposure: backtest stats, strategy gates, scans". It communicates over Streamable HTTP using the 2025-06-18 protocol revision, and does not require authorization to connect. MCPGrade currently rates Apex (smartapex.uk) D — the assessed surface is high-risk and should be remediated before use. Its most notable findings include "Tool input schema requests a secret" and "Unguarded destructive tools". This report is a deterministic, reproducible automated assessment: the same observed surface always yields the same grade under a given algorithm version, and it is refreshed as new scans arrive and free to read — but it is not a substitute for a manual human security review. Always re-check Apex (smartapex.uk)'s advertised tools, transport security, and authorization posture before connecting an autonomous agent or sharing sensitive context with it.
Add Apex (smartapex.uk) to an MCP client using the endpoint below. Review the grade and findings above before granting an autonomous agent access.
MCP endpoint
https://smartapex.uk/api/mcpStreamable HTTP transport. This is the MCP endpoint, not a website — paste it into your MCP client server list rather than a browser.
Captured passively during the read-only scan. Click any value to find servers that match it.
The tools this server advertises via tools/list — names, purposes, and the parameters each accepts, exactly as enumerated read-only.
Find Apex component cards by query, capability, tags, tier, or interface. Public reference data only; not an instruction hierarchy.
Read one Apex component card plus its verification report. Public card contract only; raw source and source packages are never returned.
Sorted worst-first. Each finding shows its severity, what it means, its OWASP MCP Top-10 mapping, and a recommended fix — the check id links to the exact methodology row that produced it.
tool: apex_run_agent_release_readiness_gate
The input schema declares a parameter that asks the caller to hand over a password, key, token, or other secret — a credential-harvest / token-passthrough risk. (Bare pagination `token` params are excluded.)
parameter "secret_scan"RecommendationNever pass secrets as tool arguments; authenticate out-of-band and remove credential parameters from the schema.
tool: apex_run_agent_dockerfile_lint
Tools that execute commands / write files / make arbitrary requests are exposed without annotations or guards.
ases, curl-pipe-shell, baked-in secrets, root user, ADD misuse, SSH eRecommendationAnnotate destructive tools and require human approval.
tool: apex_submit_review
A tool exposes a free-text/webhook/feedback parameter with no honest purpose — a classic exfil channel.
. Never include secrets, raw input, raw output, source, credentials, waRecommendationRemove side-channel parameters; constrain tool inputs.
tool: (server instructions)
A tool exposes a free-text/webhook/feedback parameter with no honest purpose — a classic exfil channel.
apex_run_agent_secret_scanner, apex_run_agent_release_readiness_gate,RecommendationRemove side-channel parameters; constrain tool inputs.
tool: (server instructions)
A tool description tries to alter the model’s use of another tool.
Discover more: apex_query_cards {"q":...} -> apex_get_card {"card_id":...} -> fRecommendationDescriptions must describe only their own tool.
The server accepts tool enumeration (and likely invocation) with no authentication.
RecommendationRequire OAuth 2.1 authorization for any server exposing non-public tools.
tool: apex_run_agent_release_readiness_gate
One tool both accesses the filesystem and reaches the network. Combining two capability classes in a single tool widens its blast radius and is the substrate for confused-deputy and exfiltration abuse.
oys, publishes, reads files, or calls CI. {"type":"object","properties":{"cRecommendationSeparate filesystem and network capabilities into distinct, independently-scoped tools.
tool: (server instructions)
The description pairs a fetch imperative with a hardcoded external URL, letting the server relocate its real instructions off-metadata and mutate them after review. Fires only when a fetch verb and a URL co-occur, so benign documentation links do not.
Missions: read https://smartapex.uk/ai-missions.json, choose one mission, and follow only its real reRecommendationDo not direct the model to fetch and act on external URLs; treat linked content as untrusted.
Financial data service providing fundamental information for 30,000+ listed companies across the US, Japan, and Korea, sourced directly from SEC, EDINET, and DART regulatory filings.
Search tours, attraction tickets, and holiday packages across 24 countries.
Agentic visitor analytics with five tools for CRO and session analysis.
52 paid x402 API endpoints for AI agents — crypto, data, DeFi, and market intelligence, settled on Base.
Access Vonage API documentation, code snippets, tutorials, and troubleshooting resources.
Run the allowlisted permissionless bounded deflated-sharpe diagnostic. DATA ONLY, read-only, no HMAC required, no source release, no orders, no wallet/account access.
Run the allowlisted permissionless bounded empyrical-style metrics calculator over supplied returns. Pure TS reimplementation of OSS empyrical 0.5.5 metric conventions. DATA ONLY, read-only, no HMAC r
Run the allowlisted permissionless bounded alphalens-style factor research wrapper over supplied factor rows. Pure TS reimplementation of selected OSS alphalens-reloaded 0.4.5 routines. DATA ONLY, rea
Run the allowlisted permissionless bounded backtrader-style single-asset backtest engine over supplied OHLCV bars and declarative strategy rules. Pure TS reimplementation of a narrow OSS backtrader 1.
Run the allowlisted permissionless bounded pandas_market_calendars 5.4.0 calendar wrapper over supplied calendar/date parameters. DATA ONLY, read-only, no HMAC required, no network, no orders, no wall
Run the allowlisted permissionless bounded pyfolio-reloaded 0.9.9 tear-sheet core over supplied dated returns. DATA ONLY, read-only, no HMAC required, no plots, no network, no orders, no wallet/accoun
Advisory exit check for stagnant positions: returns HOLD or EXIT_TIMEOUT from seconds_open and max favorable excursion. Pure calculation, DATA ONLY, read-only, no HMAC required, no orders, no wallet/a
Advisory market-regime gate: returns ALLOW or BLOCK plus move_pct from current vs reference price and a drop threshold. Pure calculation, DATA ONLY, read-only, no HMAC required, no orders, no wallet/a
Pure regime classifier over supplied prices: returns Hurst estimate and mean_reverting/random_walk/trending label. DATA ONLY, read-only, no HMAC required, no network, no orders, no wallet/account acce
Composite advisory gate over supplied statistics: CI lower bound, effect size, slippage improvement, sample depth, and Holm/BH corrected p-values. DATA ONLY, read-only, no HMAC required, no orders.
Advisory fractional Kelly sizing math from supplied win rate and average win/loss facts. DATA ONLY, read-only, no HMAC required, no orders, no wallet/account access. Not financial advice.
Run the allowlisted permissionless bounded secret scanner over supplied text/file content. DATA ONLY, read-only, no HMAC required; returns redacted findings, hashes, and line numbers only, never raw s
Run the allowlisted permissionless bounded release checklist gate. DATA ONLY, read-only, no HMAC required; evaluates caller-supplied booleans/evidence and never deploys, publishes, reads files, or cal
Run the allowlisted permissionless bounded API contract diff over supplied before/after endpoint lists. DATA ONLY, read-only, no HMAC required; no route discovery, source reads, or network access.
Advisory cooldown and size-penalty hints from a recent loss streak: returns size_multiplier, cooldown_sec, and severity. Pure calculation, DATA ONLY, read-only, no HMAC required, no orders, no wallet/
Classify supplied dependency licenses into PASS/REVIEW/BLOCK buckets using SPDX-style identifiers. DATA ONLY, read-only, no HMAC required; embedded policy table only, no registry lookups or network ac
Profile supplied tabular rows (shape, nulls, types, numeric ranges, sensitive-name hints) without echoing raw values. DATA ONLY, read-only, no HMAC required, no network access.
Redact and summarize supplied logs into severity, repeated failure patterns, and short safe samples; secret values never leave the redactor. DATA ONLY, read-only, no HMAC required, no network access.
Static best-practice lint over supplied Dockerfile text: unpinned bases, curl-pipe-shell, baked-in secrets, root user, ADD misuse, SSH exposure, apt hygiene, missing HEALTHCHECK. DATA ONLY, read-only,
Sort supplied version tags ('v1.9.0' or '1.9.0') in NUMERIC version order (v1.9.0 before v1.70.0 — which a lexical sort gets wrong). TypeScript port of a sha-pinned Yeongsu (영수) reference delivery; ma
Query the free bounded stillme dataset sample layer. Whitelisted export datasets only; no raw DB rows, no live DB connection, no trading signal. For larger slices and data receipts, use POST /v1/datas
Submit a safe usage review tied to an existing Apex verification receipt. Permissionless bounded receipts need only the receipt; verified receipts require the matching agent_id. Never include secrets,