← The evidence ledger
Evidence row 725989
Observed , by probe.mcp/1.
- Kind
- mcp.tools
- About
- endpoint efb5a9cd-32b1-4583-971f-37342439155d
- Recorded
- Previous row
- None: the first of its kind about this subject
- Body SHA-256
- eec17d8bc56a471e3b38a09be47e4b1d1c939128cd75a4a734bb61f80dbdbe85
- Leaf
- 0x87f37267db86f8226747706beace064816d4907e0978834d166e5838edd9a091
What was observed
{
"count": 30,
"tools": [
{
"description": "Analyse a set of LLM responses generated from the same prompt template but with different demographic variants (gender, origin, age, tone). Returns a bias score (0-100), sentiment analysis per variant, pairwise Jaccard similarity, and a human-readable verdict. No API key needed — runs entirely locally.",
"inputSchema": {
"properties": {
"responses": {
"description": "Array of variant responses to compare for bias",
"items": {
"properties": {
"answer": {
"description": "The LLM response text for this variant",
"type": "string"
},
"variantId": {
"description": "Identifier for the demographic variant (e.g. \"male\", \"female\", \"western\", \"young\")",
"type": "string"
}
},
"required": [
"variantId",
"answer"
],
"type": "object"
},
"minItems": 2,
"type": "array"
}
},
"required": [
"responses"
],
"type": "object"
},
"name": "bias_detect"
},
{
"description": "Audit the security attributes of cookies set by any URL. Fetches the URL and inspects all Set-Cookie headers for: HttpOnly, Secure, SameSite, Domain scope, Path scope, Max-Age/Expires, __Host-/__Secure- prefixes. Flags insecure patterns: missing HttpOnly on session cookies, missing Secure flag, SameSite=None without Secure, overly broad Domain, and excessive TTL. Returns per-cookie grades and an overall security score (0–100) — the score is the WEAKEST cookie, not the average, so one leaking session cookie cannot be averaged into a green result (average_score is reported separately).",
"inputSchema": {
"properties": {
"url": {
"description": "Full URL to audit (e.g. https://example.com/login)",
"type": "string"
}
},
"required": [
"url"
],
"type": "object"
},
"name": "cookie_security_audit"
},
{
"description": "Check the CORS configuration of a URL the same way a browser would. Returns the main response status, all Access-Control-* headers, the tested origin, and the preflight OPTIONS response. Use this for direct CORS debugging, not just security auditing.",
"inputSchema": {
"properties": {
"method": {
"description": "HTTP method to simulate (default: GET)",
"type": "string"
},
"origin": {
"description": "Origin header to simulate (default: https://yourdomain.com)",
"type": "string"
},
"url": {
"description": "Full URL to test, e.g. https://api.example.com/resource",
"type": "string"
}
},
"required": [
"url"
],
"type": "object"
},
"name": "cors_checker"
},
{
"description": "Test a URL for CORS misconfigurations. Sends preflight (OPTIONS) requests with various Origin headers to detect: wildcard origins with credentials, origin reflection (echoing any origin), null origin acceptance, subdomain wildcard bypass, and missing Vary headers. Returns risk level (safe/low/medium/high/critical) plus per-origin results. \"unknown\" means nothing was actually tested — every origin either failed to connect or answered 5xx, so the target returned no CORS decision; never read it as \"safe\". A 4xx preflight IS a real result (the server refused it and a browser would fail closed).",
"inputSchema": {
"properties": {
"origin": {
"description": "Custom Origin header to test (default: tests multiple origins automatically)",
"type": "string"
},
"url": {
"description": "Full URL to test (e.g. https://api.example.com/endpoint)",
"type": "string"
}
},
"required": [
"url"
],
"type": "object"
},
"name": "cors_test"
},
{
"description": "Decode a JWT (JSON Web Token) and return its header and payload without verifying the signature. Also reports whether the token is expired and the exact expiry date. Use to inspect claims (sub, iss, exp, roles) during debugging or when integrating with an auth provider.",
"inputSchema": {
"properties": {
"token": {
"description": "The JWT string to decode (header.payload.signature)",
"type": "string"
}
},
"required": [
"token"
],
"type": "object"
},
"name": "decode_jwt"
},
{
"description": "Scan code or config files for hardcoded secrets: AWS keys, GitHub tokens, OpenAI/Anthropic API keys, Stripe secrets, JWTs, database connection strings, and generic passwords. Returns findings with severity. Run before every commit.",
"inputSchema": {
"properties": {
"filename": {
"description": "Optional filename for context (e.g. \".env\", \"config.js\")",
"type": "string"
},
"input": {
"description": "Code or config content to scan (max 500KB)",
"type": "string"
}
},
"required": [
"input"
],
"type": "object"
},
"name": "detect_secrets"
},
{
"description": "Search available MCP tools by keyword or category before calling them. Returns matching tool names, descriptions, and optionally their inputSchemas. Call this when you are unsure which tool to use or want to explore the catalogue. Categories: data, encoding, text, llm, qa, rag, dev, security, web.",
"inputSchema": {
"properties": {
"category": {
"description": "Optional: filter by category — data | encoding | text | llm | qa | rag | dev | security | web",
"type": "string"
},
"max_results": {
"description": "Maximum tools to return (default 10, max 50). Results are ranked by IDF-weighted relevance, so common words like \"test\" do not inflate the list.",
"type": "number"
},
"query": {
"description": "Keyword(s) to search in tool name and description (e.g. \"cors\", \"token\", \"vector\", \"json\")",
"type": "string"
},
"with_schema": {
"description": "Set true to include inputSchema in results (default: false)",
"type": "boolean"
}
},
"required": [
"query"
],
"type": "object"
},
"name": "find_tool"
},
{
"description": "Validate an LLM function call / tool_use output: check that function name is in allowed list, arguments match expected schema (types, enums, ranges), no extra/missing args. Reads the definitions as OpenAI (`parameters`), MCP (`inputSchema`) or Anthropic (`input_schema`) write them, or as plain `required_args`/`optional_args`; reads the call as `arguments` (object or JSON string), `input` (Anthropic tool_use) or an OpenAI `{function:{…}}` wrapper.",
"inputSchema": {
"properties": {
"allowed_functions": {
"description": "List of allowed function definitions, in OpenAI, MCP or Anthropic form, or as {name, required_args, optional_args}",
"items": {
"properties": {
"inputSchema": {
"additionalProperties": true,
"description": "MCP form",
"type": "object"
},
"input_schema": {
"additionalProperties": true,
"description": "Anthropic form",
"type": "object"
},
"name": {
"type": "string"
},
"optional_args": {
"items": {
"type": "string"
},
"type": "array"
},
"parameters": {
"additionalProperties": true,
"description": "OpenAI form: a JSON Schema of the arguments",
"type": "object"
},
"required_args": {
"items": {
"type": "string"
},
"type": "array"
}
},
"type": "object"
},
"type": "array"
},
"function_call": {
"additionalProperties": true,
"description": "The function call object from LLM (e.g. { \"name\": \"get_weather\", \"arguments\": {\"city\":\"Paris\"} })",
"type": "object"
}
},
"required": [
"function_call",
"allowed_functions"
],
"type": "object"
},
"name": "function_call_validate"
},
{
"description": "Compute an HMAC signature for a message using a secret key. Supports SHA-256 (default), SHA-512, SHA-1, and MD5. Used for API request signing, webhook verification (GitHub, Stripe, Twilio), and JWT validation.",
"inputSchema": {
"properties": {
"algorithm": {
"description": "Hash algorithm: sha256 (default), sha512, sha1, md5",
"type": "string"
},
"encoding": {
"description": "Output encoding (default: hex)",
"enum": [
"hex",
"base64",
"base64url"
],
"type": "string"
},
"message": {
"description": "Message to sign",
"type": "string"
},
"secret": {
"description": "Secret key",
"type": "string"
}
},
"required": [
"message",
"secret"
],
"type": "object"
},
"name": "generate_hmac"
},
{
"description": "Generate a cryptographically secure random password using crypto.randomBytes. Configurable length (4–128), uppercase letters, digits, and symbols. Use when resetting user passwords, seeding test accounts, or generating API secrets.",
"inputSchema": {
"properties": {
"length": {
"description": "Password length (4–128, default: 16)",
"type": "number"
},
"numbers": {
"description": "Include digits (default: true)",
"type": "boolean"
},
"symbols": {
"description": "Include symbols like !@#$ (default: false)",
"type": "boolean"
},
"uppercase": {
"description": "Include uppercase letters (default: true)",
"type": "boolean"
}
},
"type": "object"
},
"name": "generate_password"
},
{
"description": "Test an LLM response against a set of guardrail rules: must-include, must-not-include, max length, required format, language, forbidden patterns, and custom regex. Returns pass/fail per rule.",
"inputSchema": {
"properties": {
"response": {
"description": "The LLM response to test",
"type": "string"
},
"rules": {
"description": "Array of guardrail rules to check",
"items": {
"properties": {
"label": {
"description": "Optional human-readable label",
"type": "string"
},
"type": {
"enum": [
"must_include",
"must_not_include",
"max_length",
"min_length",
"format",
"regex_match",
"regex_not_match",
"starts_with",
"ends_with",
"word_count_max",
"word_count_min"
],
"type": "string"
},
"value": {
"description": "Value for the rule (text, number as string, regex, or format name)",
"type": "string"
}
},
"type": "object"
},
"type": "array"
}
},
"required": [
"response",
"rules"
],
"type": "object"
},
"name": "guardrail_test"
},
{
"description": "Compute a cryptographic hash of a text string. Use when you need to verify data integrity, generate content fingerprints, hash passwords (prefer SHA-256+), or produce a fixed-length digest of any input. Supports SHA-256 (default), SHA-512, SHA-1, and MD5.",
"inputSchema": {
"properties": {
"algorithm": {
"description": "Hash algorithm: sha256 (default), sha512, sha1, md5",
"type": "string"
},
"input": {
"description": "Text to hash",
"type": "string"
}
},
"required": [
"input"
],
"type": "object"
},
"name": "hash_text"
},
{
"description": "Look up detailed information about any HTTP status code: class, name, description, cacheability, typical causes, and handling best practices. Covers every code in the IANA HTTP Status Code Registry (1xx-5xx, including 226, 425, 451, 508, 511 and the WebDAV codes) with its defining RFC; anything outside the registry is reported as `registered: false` rather than described. `cacheable` means heuristically cacheable by default per RFC 9110 §15.1.",
"inputSchema": {
"properties": {
"code": {
"description": "HTTP status code (e.g. 200, 404, 429, 503)",
"type": "number"
}
},
"required": [
"code"
],
"type": "object"
},
"name": "http_status_lookup"
},
{
"description": "Measure response time of one or more HTTP endpoints (GET/POST). Runs N iterations and returns min/max/avg/p95 latency. Useful for API and MCP server benchmarking.",
"inputSchema": {
"properties": {
"endpoints": {
"description": "Endpoints to benchmark. Accepts a single URL string, an array of URL strings, or an array of {url, method?, body?, headers?, label?} objects.",
"items": {
"properties": {
"body": {
"additionalProperties": true,
"description": "Request body for POST",
"type": "object"
},
"headers": {
"additionalProperties": true,
"description": "Custom headers",
"type": "object"
},
"label": {
"description": "Optional label for this endpoint",
"type": "string"
},
"method": {
"description": "HTTP method (default: GET)",
"enum": [
"GET",
"POST"
],
"type": "string"
},
"url": {
"description": "Full URL to test",
"type": "string"
}
},
"type": "object"
},
"type": [
"string",
"array"
]
},
"iterations": {
"description": "Number of iterations per endpoint (default: 3, max: 10)",
"type": "number"
}
},
"required": [
"endpoints"
],
"type": "object"
},
"name": "latency_benchmark"
},
{
"description": "Lint an MCP tool definition for best practices: naming conventions, description quality, schema completeness, required fields consistency, description length. Returns actionable warnings.",
"inputSchema": {
"properties": {
"tool_definition": {
"additionalProperties": true,
"description": "MCP tool definition object with name, description, inputSchema",
"type": "object"
}
},
"required": [
"tool_definition"
],
"type": "object"
},
"name": "mcp_schema_lint"
},
{
"description": "Run a compliance evaluation against a live MCP server over Streamable HTTP, the way a client meets it. Checks: initialize (protocol version, serverInfo, Mcp-Session-Id carried afterwards), notifications/initialized (202), tools/list with pagination, the spec's tool-definition rules, ping, an unknown method (-32601), one tools/call, GET on the endpoint (SSE or 405), and P50/P95 latency over pings. JSON and SSE answers are both read. Only a tool you name in test_tool_name, or one the server annotates readOnlyHint: true, is called — otherwise the call is skipped and not scored. Returns PASS/FIX/BLOCK with a 0-100 score and per-check details; INCONCLUSIVE (score null) when the server requires authorization, since no credentials are sent. A 301/302 on POST caps the verdict at FIX. The call runs from ia-qa.com: private and local addresses are refused, and stdio servers cannot be reached — for those, paste their tools/list answer into mcp_server_health_check. structuredContent is checked for presence, not validated against outputSchema.",
"inputSchema": {
"properties": {
"test_tool_name": {
"description": "Tool to call in the tools/call step. Without it, the first tool annotated readOnlyHint: true is called, or none.",
"type": "string"
},
"url": {
"description": "Public URL of the MCP endpoint (e.g. https://mcp.example.com/mcp). An origin alone (https://www.ia-qa.com) tries /mcp, then /.",
"type": "string"
}
},
"required": [
"url"
],
"type": "object"
},
"name": "mcp_server_evaluate"
},
{
"description": "Check an MCP server's tool definitions without contacting it — for a server not deployed yet, a stdio server, or a CI gate. Paste the answer to tools/list (raw JSON-RPC or its result), a bare array of tools, one tool definition, or a manifest carrying name/description; input_shape says which one was read. Validates every tool against the MCP spec (name 1-128 chars of A-Z a-z 0-9 _ - .; camelCase and a tool with no parameters are valid; inputSchema of type object; required fields listed in properties) and, with strict, documentation (description ≥ 20 chars, every parameter with a description and a type). verdict: FAIL when a tool breaks a rule or no tool is found, WARN when only a manifest's server name/description is missing, PASS otherwise. A tools/list carries no server name, so those two checks are not scored on it (the name comes from initialize). Gate on verdict or on checks[tool_quality].pass.",
"inputSchema": {
"properties": {
"manifest": {
"description": "JSON text: the answer to tools/list (raw JSON-RPC or its result), an array of tools, one tool definition, or a manifest with name/description/tools.",
"type": "string"
},
"strict": {
"description": "Also require documentation: every description ≥ 20 characters, every parameter with a description and a type.",
"type": "boolean"
}
},
"required": [
"manifest"
],
"type": "object"
},
"name": "mcp_server_health_check"
},
{
"description": "Parse a raw HTTP headers block into a structured JSON object. Detects multi-value headers, masks Authorization values, and optionally audits for missing security headers (HSTS, CSP, X-Frame-Options, etc.).",
"inputSchema": {
"properties": {
"analyze_security": {
"description": "Audit for missing security headers (default: true)",
"type": "boolean"
},
"headers": {
"description": "Raw HTTP headers (one \"Name: Value\" per line)",
"type": "string"
}
},
"required": [
"headers"
],
"type": "object"
},
"name": "parse_http_headers"
},
{
"description": "Scan user input or prompts for common prompt injection patterns. Pattern-based: it recognises known phrasings and structural markers, not intent — a paraphrase it has no pattern for, or a language it does not cover, reads clean, so a clean result is not proof of safety. The instruction-override phrasing is covered in English, French, Spanish, German, Italian, Portuguese, Dutch, Russian, Chinese, Japanese and Arabic, and the text is also matched after removing invisible characters, lookalike letters, spaced-out letters and leetspeak (such a hit is marked `obfuscated`). Detects system prompt overrides, jailbreak attempts, role manipulation, encoding tricks, delimiter attacks (the control tokens of ChatML and Llama chat templates, and fake role headers that imitate a markdown heading or a dashed separator announcing a new system turn), template/interpolation injection ({{...}}, ${...}), and context-exfiltration attempts (asking the model to repeat the text that precedes the request). A match inside quoted or fenced text (documentation citing a payload) is reported one severity level lower and marked `quoted` — never suppressed, since an LLM reading the document as data can still follow a quoted instruction. A quote preceded by a live imperative (\"output the following: ...\") keeps its full severity.",
"inputSchema": {
"properties": {
"input": {
"description": "The user input or prompt to scan for injection patterns",
"type": "string"
},
"sensitivity": {
"description": "Detection sensitivity (default: medium)",
"enum": [
"low",
"medium",
"high"
],
"type": "string"
}
},
"required": [
"input"
],
"type": "object"
},
"name": "prompt_injection_scan"
},
{
"description": "Give honest usage feedback on an IA-QA MCP tool. Provide a score (1-5) and a comment. Rate low (1-2) if the tool was wrong, irrelevant, or a poor fit; rate high (4-5) only if it genuinely solved your need. Ratings are read by the IA-QA maintainers and are not published, one vote per agent and tool a month. Comments are screened before storing: one carrying a credential is not stored and personal data is replaced by placeholders — describe what happened rather than what a model should do. Skip rating routine successes — we want signal, not praise. Example: rate_tool({ tool_name: \"format_json\", score: 2, comment: \"Tried to pretty-print a JSON5 file, it rejected trailing commas — not usable for my case.\" })",
"inputSchema": {
"properties": {
"comment": {
"description": "Strongly encouraged — explain what you were trying to do and whether the tool got you there. Be specific about what was missing, wrong, or a poor fit. This is the most valuable part of the rating. Up to 2000 chars are stored; go over and the response says so (truncated: true) — send the remainder as a second call rather than assuming it landed.",
"type": "string"
},
"score": {
"description": "Rating from 1 (poor) to 5 (excellent)",
"maximum": 5,
"minimum": 1,
"type": "number"
},
"tool_name": {
"description": "Name of the MCP tool to rate (e.g. \"format_json\", \"shield_analyze\")",
"type": "string"
}
},
"required": [
"tool_name",
"score"
],
"type": "object"
},
"name": "rate_tool"
},
{
"description": "Automatically detect and redact Personally Identifiable Information (PII) from text. Replaces emails, phone numbers, SSNs, IBANs, credit cards, IP addresses, and JWT tokens with [REDACTED_TYPE] placeholders. Safe to use before logging or sending to an LLM.",
"inputSchema": {
"properties": {
"input": {
"description": "Text to redact PII from",
"type": "string"
},
"marker": {
"description": "Custom replacement marker (default: \"REDACTED\"). Result: [REDACTED_EMAIL]",
"type": "string"
},
"types": {
"description": "Comma-separated types to redact (default: all). Options: email, phone, ssn, iban, credit_card, ip_address, jwt",
"type": "string"
}
},
"required": [
"input"
],
"type": "object"
},
"name": "redact_pii"
},
{
"description": "Scan text or code for leaked secrets: API keys (AWS, GCP, Azure, OpenAI, Anthropic, Stripe, GitHub, GitLab, Slack, Twilio, SendGrid, HuggingFace), private keys (RSA/EC/PGP), JWTs, database connection strings, Bearer tokens, and Basic auth headers. Returns a list of findings with type, severity, line number, and a redacted preview. Use before committing code, sharing logs, or sending text to an LLM. 100% regex-based, zero network calls.",
"inputSchema": {
"properties": {
"input": {
"description": "Text or code to scan for secrets",
"type": "string"
},
"types": {
"description": "Comma-separated families to scan (default: all): aws, gcp, azure, openai, anthropic, huggingface, github, gitlab, stripe, slack, twilio, sendgrid, jwt, private_key, connection_string, bearer, basic_auth, generic. Individual pattern names (e.g. \"aws_access_key\", \"github_fine\") are also accepted. An unknown value is rejected with an error — a scoped scan never silently returns \"clean\".",
"type": "string"
}
},
"required": [
"input"
],
"type": "object"
},
"name": "secret_scan"
},
{
"description": "Analyse the HTTP security headers of a public URL OR of raw response headers you paste in. Grades each header (A–F) for: Strict-Transport-Security, Content-Security-Policy, X-Frame-Options, X-Content-Type-Options, Referrer-Policy, Permissions-Policy, X-XSS-Protection, Cross-Origin-Opener-Policy, Cross-Origin-Resource-Policy, and Cross-Origin-Embedder-Policy. Returns an overall score (0–100), per-header grades, missing headers, and fix snippets for Express, Nginx, and Apache. For localhost/private targets the remote server cannot reach, pass the `headers` parameter instead of `url`.",
"inputSchema": {
"properties": {
"headers": {
"description": "Optional, and sufficient on its own (no url needed). The response headers to grade, either as an object {\"strict-transport-security\": \"max-age=...\", ...} or as the raw header block pasted as a string (e.g. `curl -sI` output). Use this to audit a local server the remote MCP cannot reach."
},
"url": {
"description": "Optional. Full public URL to check (e.g. https://example.com). Omit it entirely when using `headers`. The server cannot reach localhost/private IPs.",
"type": "string"
}
},
"type": "object"
},
"name": "security_headers_check"
},
{
"description": "Run a comprehensive AI guardrail analysis on an LLM response. Orchestrates 7 deterministic safety checks plus an optional LLM-powered deep analysis in parallel: hallucination detection (grounding score), prompt injection scan on BOTH the response and the original prompt (reported separately as checks.injection / checks.injection_prompt, scored once at the worse of the two), toxicity scan, output validation (PII/safety), guardrail rules, response quality scoring, and AI verdict (via Qwen, Gemma, Llama, etc.). Returns a unified PASS/FIX/BLOCK verdict with a 0-100 safety score, per-check results, and actionable fix recommendations. PII gates the verdict: an SSN or credit card in the response forces BLOCK, an email forces FIX, phone/IP matches are flagged only (their regexes also match dates and version strings). Use this as a single-call safety gate before surfacing any LLM output to users.",
"inputSchema": {
"properties": {
"model": {
"description": "LLM model for AI-powered deep analysis (default: \"openai/gpt-oss-20b\"). Set to \"none\" to skip LLM check. Supports any model from list_llm_models.",
"type": "string"
},
"prompt": {
"description": "Optional original prompt. Used for quality scoring AND scanned for prompt injection in its own right (checks.injection_prompt) — pass it whenever you have it, it is where the attack actually lands.",
"type": "string"
},
"response": {
"description": "The LLM-generated response to analyze",
"type": "string"
},
"rules": {
"description": "Optional guardrail rules array (same format as guardrail_test tool)",
"items": {
"properties": {
"label": {
"type": "string"
},
"type": {
"type": "string"
},
"value": {
"type": "string"
}
},
"type": "object"
},
"type": "array"
},
"source": {
"description": "Optional reference/source text for hallucination grounding check",
"type": "string"
}
},
"required": [
"response"
],
"type": "object"
},
"name": "shield_analyze"
},
{
"description": "Analyse the SSL/TLS certificate of any HTTPS host. Returns certificate subject, issuer, validity dates, days until expiry, protocol version, cipher suite, key exchange info, and an overall grade (A+, A, B, C, F). Detects expired, self-signed, and weak certificates. Use this to audit TLS posture before production deployment or during security reviews.",
"inputSchema": {
"properties": {
"host": {
"description": "Hostname to check (e.g. example.com). Do not include https:// prefix.",
"type": "string"
},
"port": {
"description": "Port number (default: 443)",
"type": "number"
}
},
"required": [
"host"
],
"type": "object"
},
"name": "ssl_certificate_check"
},
{
"description": "Scan text for toxic language, hate speech, bias/stereotype framing, violence, sexual and self-harm content. Lexical + structural pattern matching (identity term + predicate), not a semantic classifier — returns per-category risk plus the named rules that fired, so every finding can be checked. Languages: English for every category; French for insults, coarse language, direct threats and self-harm intent; anything else reads clean, so clean is not proof of safety. Text is also matched after undoing leetspeak and lookalike letters. Self-harm is critical only on a first-person intent — a mention (\"call 988 if you have suicidal thoughts\") is medium. Useful for LLM safety guardrail testing and triage; signal-only, not a calibrated CI gate.",
"inputSchema": {
"properties": {
"categories": {
"description": "Categories to check (default: all)",
"items": {
"enum": [
"profanity",
"hate_speech",
"violence",
"sexual",
"self_harm",
"bias"
],
"type": "string"
},
"type": "array"
},
"text": {
"description": "Text to scan",
"type": "string"
}
},
"required": [
"text"
],
"type": "object"
},
"name": "toxicity_scan"
},
{
"description": "Run declarative assertions on an agent trace (OpenAI tool-call messages, Anthropic tool_use/tool_result blocks, LangChain run trees, or plain text ReAct logs). No LLM call — deterministic. Assertion types: order (tool A before B), must_call, must_not_call, max_calls, min_calls, no_error, recovery (agent completes a successful step after its last error). A step counts as errored when the trace says so — is_error/isError, status/state in {error,failed,exception,…}, an error field, a JSON body with error/success:false — or when its text payload STARTS with an error marker (Error:, Traceback, TypeError:). Model prose is never scanned for keywords, and every errored step reports error_signal naming what flagged it. Returns per-assertion PASS/FAIL, parsed steps, warnings, and an overall verdict: PASS, FAIL, or INCONCLUSIVE when the trace parsed to 0 steps (assertions passing on emptiness prove nothing, so it is never PASS). An empty assertions list is refused. Use this to gate CI/CD on agent behavior correctness.",
"inputSchema": {
"properties": {
"assertions": {
"description": "List of assertions to validate against the trace.",
"items": {
"properties": {
"after": {
"description": "[order] Tool that must be called after.",
"type": "string"
},
"before": {
"description": "[order] Tool that must be called first.",
"type": "string"
},
"id": {
"description": "Optional assertion identifier.",
"type": "string"
},
"max": {
"description": "[max_calls] Maximum number of allowed calls.",
"type": "number"
},
"min": {
"description": "[min_calls] Minimum number of required calls.",
"type": "number"
},
"sensitivity": {
"description": "[no_injection] Default low = critical patterns only (override, jailbreak, fake role delimiters). Higher catches more and fires on ordinary text (\"act as\", {{templates}}) — pattern matching, not a classifier.",
"enum": [
"low",
"medium",
"high"
],
"type": "string"
},
"tool": {
"description": "Tool name to check (for must_call, must_not_call, max_calls, min_calls).",
"type": "string"
},
"type": {
"description": "Assertion type. no_injection runs prompt_injection_scan on every tool result (indirect injection: text a third party wrote that the agent then reads); no_pii runs redact_pii on the final answer (a leak to the user). Both report types and counts, never the matched values.",
"enum": [
"order",
"must_call",
"must_not_call",
"max_calls",
"min_calls",
"no_error",
"recovery",
"no_injection",
"no_pii"
],
"type": "string"
},
"types": {
"description": "[no_pii] Comma-separated PII types (email, phone, ssn, credit_card, ip_address, jwt). Default: all.",
"type": "string"
},
"where": {
"description": "[no_injection, no_pii] What to scan. Defaults: no_injection → tool_results, no_pii → answer (PII in a tool result is often legitimate — an order carries the customer email).",
"enum": [
"tool_results",
"answer"
],
"type": "string"
}
},
"required": [
"type"
],
"type": "object"
},
"maxItems": 30,
"type": "array"
},
"format": {
"description": "Trace format. auto (default) detects automatically.",
"enum": [
"auto",
"openai",
"langchain"
],
"type": "string"
},
"trace": {
"description": "Agent execution trace as JSON (OpenAI messages array, LangChain run tree) or plain text log (Thought/Action/Observation format).",
"oneOf": [
{
"type": "string"
},
{
"additionalProperties": true,
"type": "object"
},
{
"items": {},
"type": "array"
}
]
}
},
"required": [
"trace",
"assertions"
],
"type": "object"
},
"name": "validate_agent_trajectory"
},
{
"description": "Validate that an MCP tool response conforms to expected format, schema, and content rules. Use this to QA-test any MCP server tool. Supply the tool's actual JSON result and a set of checks to perform.",
"inputSchema": {
"properties": {
"actual_latency": {
"description": "Actual measured latency in ms (from the call)",
"type": "number"
},
"expected_type": {
"description": "Expected top-level type: \"object\", \"array\", \"string\", \"number\"",
"enum": [
"object",
"array",
"string",
"number"
],
"type": "string"
},
"forbidden_keys": {
"description": "Comma-separated list of keys that MUST NOT exist (e.g. \"password, secret, token\")",
"type": "string"
},
"max_response_ms": {
"description": "Maximum acceptable latency in ms (will be compared if provided)",
"type": "number"
},
"max_size_bytes": {
"description": "Maximum acceptable response size in bytes",
"type": "number"
},
"min_items": {
"description": "If response is an array, minimum number of items expected",
"type": "number"
},
"required_keys": {
"description": "Comma-separated list of keys that MUST exist in the response (dot-notation for nested: \"data.id, data.name\")",
"type": "string"
},
"response": {
"description": "The MCP tool result as a JSON string to validate",
"type": "string"
}
},
"required": [
"response"
],
"type": "object"
},
"name": "validate_mcp_response"
},
{
"description": "Parse and validate a URL. Returns decomposed components: protocol, hostname, port, path, query parameters, hash, and origin.",
"inputSchema": {
"properties": {
"input": {
"description": "URL to validate and parse",
"type": "string"
}
},
"required": [
"input"
],
"type": "object"
},
"name": "validate_url"
},
{
"description": "Run a comprehensive web security audit combining headers, SSL, CORS, and cookies checks — then use an LLM to produce a prioritised remediation plan. Orchestrates security_headers_check + ssl_certificate_check + cors_test + cookie_security_audit in parallel, merges all findings, then asks an AI model to: (1) rank vulnerabilities by real-world exploitability, (2) generate a remediation roadmap, (3) produce fix code snippets for the detected stack. Returns both raw audit data and the AI analysis. Use this as a one-click security posture assessment.",
"inputSchema": {
"properties": {
"api_key": {
"description": "Your Groq or HuggingFace API key. Required to enable AI analysis.",
"type": "string"
},
"model": {
"description": "LLM model for AI analysis (default: \"openai/gpt-oss-20b\"). Set to \"none\" to skip AI analysis.",
"type": "string"
},
"url": {
"description": "Full URL to audit (e.g. https://example.com)",
"type": "string"
}
},
"required": [
"url"
],
"type": "object"
},
"name": "web_security_audit"
}
]
}What the leaf hashes
{"body":"eec17d8bc56a471e3b38a09be47e4b1d1c939128cd75a4a734bb61f80dbdbe85","id":725989,"kind":"mcp.tools","method":"probe.mcp/1","observed_us":1793150763340704,"prev":null,"subject":"efb5a9cd-32b1-4583-971f-37342439155d","subject_kind":"endpoint"}Proof
Fold these into the leaf in order, hashing each sorted pair with keccak256, to reach the root of (231336 rows).
- 0x87f384961e16dfd7c7e2a092acb3e01ea8cd1dc23735c70aa552175607a69d45
- 0xb10a1e845a7f986337fb7d75979672edea5f47ae567a522eb1f76dec28992311
- 0xb1b0ad44c3c85142c34775cca71b9053cd81fe1ac40039bc450f4fff9ed1118c
- 0x76d651d1b9bb74a3401258f7a5ba52f2054a48e51559964fda28d51f66b01739
- 0x4a96aa0565e08ef9a8597aedf27d814ba91dba47ad1963856dad43620ce781e7
- 0x61e4a446430cef03a394c19182ad255d25855d705604e01bc52d1ea44a8953e1
- 0x716116e15cfb079e088311e362b29991f1b8198432b164bc23d17c9118b3479f
- 0xedfbae6232e8e834b510beec6a7510e47bdf487d18d1b5bce8f3b0fe546ef39f
- 0x2abdac483b7c26ef29746cfcab4ca42bb4bb8aebdafa8438401ff22078847744
- 0x0944c3bae565949a335daca8abf7cd78ad83da998c47ad188e9c1f9e935b64ec
- 0xfcb87fd1f97629d9bbe729e1af67ec8754514bfe9ee1cc45cf69c5e026b58ce0
- 0x1810cd44ebfe5f26ef4925cdfbe26287c4f9dbf1cd3343ea6968181a15656337
- 0x73b20f1c980fc42d6a2767acddce2cb41764e0b54fa3342f0dc75bd609f6023e
- 0x115e196b75a0422533397227b51b9c807591ec2574919c03800bb0d253f110e8
- 0xe2f9c120d14f36fdddf9b4e605c5d0b8a13b835e2380472255dc22e14b93e84e
- 0x74cf00e07b52415c2184cf20be69d195ed2725497a6892c2f3e449497c17d0ec
- 0x7a743a0a09057bae687051274e1d45bf973bd36a1a4ba16a191b5d29e3bdfa2a
- 0xe027395fafb7fc41921c42ca0542524ea22204cbd5908b70b27f3d5fb26f5d0f
- Day root
- 0x2c19037464cc4c493624565f01775d6446fc78eaa118fc6865bfe2bfca9ff78e
- Chained
- 0xd1d7dee8322ceff8f4c8d83aa1fe4a4466825c2af1c97dbde3d35db61234f7be
- Published
- On Arc Testnet, block 69584042: 0x8bae2b1d097834c8b458513862d70d29ee807e98dc2e863b27250131c15b35ae