Loading ToolBench

Reading the current corpus…

mcp-for-security MCP server · F · ToolBench

mcp-for-security

github.com/hackerdogs-ai/mcp-for-security

A comprehensive MCP server providing integration with multiple security testing and reconnaissance tools including Alterx, Amass, Arjun, AssetFinder, Cero, Commix, crt.sh, ffuf, GoWitness, HTTP Headers Security, httpx, Katana, Masscan, MobSF, Nmap, Nuclei, ScoutSuite, ShuffleDNS, and Smuggler

2026-07-28+STDIO1 tools

45/100

Definition Quality45

Protocol Readiness50

Supportability41

Spec posture

Inferred effective spec 2026-07-28+·current baseline 2026-07-28

Static source inference · medium confidence

Current-spec patterns detected

Summary

Single tool 'do-smuggler' has a reasonable name starting with verb 'do', but the description is minimal (67 chars, below 10-char baseline). Input schema is present with proper types (string, array) and parameter descriptions, but lacks critical details: no enum constraints for smuggler_args, no output schema documented, no error handling guidance. The tool description does not explain WHEN to use it, WHAT it returns, or how to interpret results. Parameter descriptions are present but generic, 'Additional smuggler arguments' with a list of flags is not actionable for an LLM without explaining what each flag does or when to use them. No mention of prerequisites, dependencies, or expected output format. Follows minimal MCP structure but falls short of production-grade tool design.

Tools (1)

do-smugglerread onlysource verified45/100

Run Smuggler to detect HTTP Request Smuggling vulnerabilities

Issues (5)

CRITICALtool-description ↗

Tool description is too brief (67 chars) and lacks actionable context. Does not explain WHEN to use the tool, WHAT it returns, or how to interpret results. LLMs cannot determine when to select this tool over alternatives.

do-smuggler

CRITICALtool ↗

No output schema documented. LLMs cannot plan downstream actions or extract results without knowing what fields to expect. Critical for tool composition.

do-smuggler

HIGHtool-description ↗

Parameter 'smuggler_args' description lists raw CLI flags without explaining their purpose or when to use them. LLMs cannot reason about which flags to pass without semantic guidance.

do-smuggler

HIGHrecovery-guide ↗

No error handling guidance. Tool does not explain what errors are possible, how to recover, or what to do if Smuggler fails or times out.

do-smuggler

Recommendations

Expand tool description to 100-200 chars explaining: 'Detects HTTP Request Smuggling vulnerabilities in a target URL using Smuggler. Returns vulnerability findings with severity levels. Use this to identify CL.TE, TE.CL, and TE.TE smuggling vectors. Requires network access to the target.'
Document output schema explicitly: 'Returns an object with fields: vulnerabilities (array of {type, severity, payload}), summary (string), timestamp (ISO 8601). If no vulnerabilities found, returns empty vulnerabilities array.'
Replace raw CLI flags in smuggler_args description with semantic guidance: 'Optional flags to customize detection. Common: -m POST (HTTP method), -t 5 (timeout seconds), -x (exit on first finding), -verify quick (verification level). See Smuggler docs for full list.'
Add error handling section to description: 'Returns error if URL is unreachable, invalid, or times out. Retryable errors include timeout and connection refused. Non-retryable: invalid URL format, host not found.'
Add URL parameter constraint: 'Must be a valid HTTP or HTTPS URL (e.g., https://example.com). Tool will fail if URL is unreachable or does not respond within timeout.'
Consider splitting smuggler_args into named parameters (method, timeout, verify_level) for better LLM reasoning and validation.

Score history

Overall score trend

First recorded score · v2 rubric

45/100

Scored	Grade	Overall	Spec posture	Rubric
2026-09-23	F	45	2026-07-28+	v2