Skip to main content
Glama
Sign In

EdgeScreen x402 Security Screening

by dev.workers.asostudio.aso-x402-screen
OverviewDiscussionsAdmin

screen_batch

Read-onlyIdempotent

Screen up to 25 crypto addresses (OFAC SDN), domains/handles (IDN homoglyphs), untrusted prompts (prompt injection & tool poisoning), URLs (pre-fetch SSRF firewall), text snippets (secret leak redaction), and/or git diffs (pre-merge patch security) in <1ms. Requires $0.005 (5000 atomic) Base USDC on eip155:8453 via x402 PAYMENT-SIGNATURE header.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "screen_batch"
  }
}

Input Schema

NameRequiredDescriptionDefault
urlNoSingle URL shorthand (up to 2048 chars) to inspect for SSRF, private/metadata IPs, dangerous schemes, credentials, IDN homoglyphs, and open redirects.
diffNoSingle unified git diff or PR patch shorthand (up to 16384 chars) to audit for leaked secrets, destructive shell/SQL, CI injection, malicious lifecycle hooks, SSRF/private IPs, auth/TLS bypass, Trojan Source Unicode, and corrupt hunk counts.
textNoAlias for `secret`: single text/log/PR snippet (up to 8192 chars) to scan & redact for leaked secrets.
urlsNoURLs (up to 2048 chars each) to pre-screen before fetch for SSRF (loopback, RFC1918, CGNAT, IPv6 mapped/local, cloud metadata, hex/octal/decimal IP encodings), non-HTTP(S) schemes, credentials, punycode homoglyphs, and open redirects.
brandNoTarget brand for deterministic homoglyph, Punycode, and typosquatting comparison.
diffsNoUnified git diffs or PR patches (up to 16384 chars each) to pre-screen before git apply / PR merge.
patchNoAlias for `diff`: single unified git diff or patch snippet (up to 16384 chars) to audit.
textsNoAlias for `secrets`: array of text snippets (up to 8192 chars each) to scan & redact for leaked secrets.
domainNoSingle candidate domain shorthand (ASCII, Unicode IDN, or Punycode xn--...) to inspect.
handleNoSingle candidate social handle shorthand (@handle or bare handle) to inspect.
promptNoSingle untrusted prompt, MCP tool description, or fetched text (up to 4096 chars) to scan.
secretNoSingle text/log/diff/webhook payload shorthand (up to 8192 chars) to scan & redact for leaked API keys, private keys, JWTs, and Bearer tokens.
addressNoSingle crypto address shorthand (EVM 0x..., Bitcoin 1.../3.../bc1..., TRON T...) to screen against OFAC SDN.
domainsNoCandidate domains (ASCII, Unicode IDN, or Punycode xn--...) to inspect.
handlesNoCandidate social handles (@handle or bare handle) to inspect.
patchesNoAlias for `diffs`: array of unified git diffs or patch snippets (up to 16384 chars each) to audit.
promptsNoUntrusted prompts, MCP tool descriptions, or fetched texts (up to 4096 chars each) to scan for prompt injection, tool poisoning, Unicode tag smuggling, and secret exfiltration.
secretsNoText snippets, PR bodies, logs, or webhook payloads (up to 8192 chars each) to scan & redact for leaked API keys (OpenAI, Anthropic, AWS, GitHub, Stripe, Slack, Google), PEM/hex private keys, JWTs, and Bearer tokens.
addressesNoCrypto addresses (EVM 0x..., Bitcoin 1.../3.../bc1..., TRON T...) to screen against OFAC SDN.
candidateNoSingle candidate domain or social handle shorthand to auto-classify and inspect.
candidatesNoMixed candidate domains or social handles to auto-classify and inspect.
officialDomainsNoOptional list of verified official domains for the target brand.
officialHandlesNoOptional list of verified official social handles for the target brand.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed4 schema fields changed
    • addedInput schema / properties / diff
      Added value: +{
+  "description": "Single unified git diff or PR patch shorthand (up to 16384 chars) to audit for leaked secrets, destructive shell/SQL, CI injection, malicious lifecycle hooks, SSRF/private IPs, auth/TLS bypass, Trojan Source Unicode, and corrupt hunk counts.",
+  "maxLength": 16384,
+  "type": "string"
+}
    • addedInput schema / properties / diffs
      Added value: +{
+  "description": "Unified git diffs or PR patches (up to 16384 chars each) to pre-screen before git apply / PR merge.",
+  "items": {
+    "maxLength": 16384,
+    "type": "string"
+  },
+  "maxItems": 25,
+  "type": "array"
+}
    • addedInput schema / properties / patch
      Added value: +{
+  "description": "Alias for `diff`: single unified git diff or patch snippet (up to 16384 chars) to audit.",
+  "maxLength": 16384,
+  "type": "string"
+}
    • addedInput schema / properties / patches
      Added value: +{
+  "description": "Alias for `diffs`: array of unified git diffs or patch snippets (up to 16384 chars each) to audit.",
+  "items": {
+    "maxLength": 16384,
+    "type": "string"
+  },
+  "maxItems": 25,
+  "type": "array"
+}
  2. First observed

TDQS

A3.6/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly, idempotent, non-destructive, closed-world behavior, so the bar is lower — yet the description adds genuinely non-obvious operational facts: sub-1ms latency and a hard payment requirement of $0.005 in Base USDC on eip155:8453 via an x402 PAYMENT-SIGNATURE header. It does not say what happens on missing or insufficient payment (error vs. 402 challenge), which is the main remaining gap.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two sentences, with the capability scope front-loaded before the payment caveat, and no filler. The enumeration is dense but each clause carries a distinct screening domain; the payment clause is appropriately placed last as a precondition.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 23-parameter tool with no required fields and no output schema, the description covers scope and payment but omits critical invocation guidance: that at least one screening input must be supplied, and how the alias groups (text/secret/texts/secrets, diff/patch/diffs/patches) relate. It also does not characterize the response shape, which matters when screening six heterogeneous input classes at once.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so every one of the 23 parameters is already documented in the input schema, including length caps, aliases, and per-category threat lists. The description adds only the high-level category-to-capability mapping, not syntax, precedence between alias groups, or how mixed inputs are handled — baseline 3 applies.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

Opens with a specific verb ("Screen") and enumerates exactly which resources are screened — crypto addresses, domains/handles, prompts, URLs, text, git diffs — with the per-category threat model in parentheses. Batch scope ("up to 25") implicitly differentiates it from the single-item siblings like screen_address and screen_diff, but it never states that relationship outright.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Usage moments are implied by phrases like "pre-fetch SSRF firewall" and "pre-merge patch security," which tell an agent where in a workflow this fits. There is no explicit when-to-use versus the single-item siblings and no when-not guidance, so routing between screen_batch and screen_url/screen_diff is left to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All EdgeScreen x402 Security Screening toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources