# Diogenes Agent Guardrails & Sandbox MCP (remote · diogenes-micro-gateway.corinth.workers.dev)

Autonomous agent guardrails, bash sandboxing, MCP rugpull diffing & prompt firewalls.

- Trust score: 65/100 (medium)
- Change this week: +4
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- remote · `diogenes-micro-gateway.corinth.workers.dev`: 65/100 (this document), [markdown](https://verifymcp.io/servers/stand-out-of-my-sun-diogenes-agent-guardrails-mcp/mcp-agent-guardrails.md), [page](https://verifymcp.io/servers/stand-out-of-my-sun-diogenes-agent-guardrails-mcp/mcp-agent-guardrails)

## Channel facts

- Endpoint: `https://diogenes-micro-gateway.corinth.workers.dev/mcp/agent-guardrails`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.0.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-05.

- **Endpoint Security**: 74/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - No authorisation is required to call this server. Every tool declares its destructiveHint and none is destructive, so open access doesn't expose one.
  - HTTPS is enforced; there's no plaintext access path.
  - HSTS check failed: the Strict-Transport-Security header is absent.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 65/100
  - AI-judged instruction clarity (excellent).
  - Context-footprint check failed: tool/resource definitions use about 2424 tokens (~202/item across 12 items; 12 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 0/100
  - Stability check failed: schema churn in the 16 days we've observed: 9 tool removals, 0 breaking changes, 0 auth/transport breaks, 12 additions.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Tool Safety**: 75/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - 0 of 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "execute_python_sandbox" implies "execute" and declares readOnlyHint instead, contradicting what its own name says it does.
  - An AI judge read all 13 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 20/100
  - Spec-recency check failed: implements MCP spec 2024-11-05; the latest is 2026-07-28.

## Install

### How do I install the Diogenes Agent Guardrails & Sandbox MCP server?

Diogenes Agent Guardrails & Sandbox MCP is a hosted endpoint at https://diogenes-micro-gateway.corinth.workers.dev/mcp/agent-guardrails, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http stand-out-of-my-sun-diogenes-agent-guardrails-mc 'https://diogenes-micro-gateway.corinth.workers.dev/mcp/agent-guardrails'
```

### Cursor

```json
{
  "mcpServers": {
    "stand-out-of-my-sun-diogenes-agent-guardrails-mc": {
      "url": "https://diogenes-micro-gateway.corinth.workers.dev/mcp/agent-guardrails"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "stand-out-of-my-sun-diogenes-agent-guardrails-mc": {
      "type": "http",
      "url": "https://diogenes-micro-gateway.corinth.workers.dev/mcp/agent-guardrails"
    }
  }
}
```

### Codex

```toml
[mcp_servers.stand-out-of-my-sun-diogenes-agent-guardrails-mc]
url = "https://diogenes-micro-gateway.corinth.workers.dev/mcp/agent-guardrails"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "stand-out-of-my-sun-diogenes-agent-guardrails-mc": {
      "type": "remote",
      "url": "https://diogenes-micro-gateway.corinth.workers.dev/mcp/agent-guardrails",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add stand-out-of-my-sun-diogenes-agent-guardrails-mc --url 'https://diogenes-micro-gateway.corinth.workers.dev/mcp/agent-guardrails' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  stand-out-of-my-sun-diogenes-agent-guardrails-mc:
    url: "https://diogenes-micro-gateway.corinth.workers.dev/mcp/agent-guardrails"
```

### Netclaw

```json
{
  "McpServers": {
    "stand-out-of-my-sun-diogenes-agent-guardrails-mc": {
      "Transport": "http",
      "Url": "https://diogenes-micro-gateway.corinth.workers.dev/mcp/agent-guardrails"
    }
  }
}
```

### Vellum

```bash
assistant mcp add stand-out-of-my-sun-diogenes-agent-guardrails-mc -t streamable-http -u 'https://diogenes-micro-gateway.corinth.workers.dev/mcp/agent-guardrails'
```

### Other

```json
{
  "mcpServers": {
    "stand-out-of-my-sun-diogenes-agent-guardrails-mc": {
      "type": "http",
      "url": "https://diogenes-micro-gateway.corinth.workers.dev/mcp/agent-guardrails"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-03 (score 65, +47)

- [security regression] Stability: unverified → fail
- [security improvement] Authorization: unverified → partial
- [security improvement] HTTPS: unverified → pass
- [security improvement] Injection markers: unverified → pass
- [security improvement] Transport: fail → pass
- [security] The server rewrote its instructions, which are the text every model session reads
- [functional regression] MCP protocol: unverified → fail
- [functional improvement] Endpoint reachability: not serving MCP → reachable
- [functional improvement] Tool coverage: unverified → 100

### 2026-11-02 (score 18, −43)

- [security regression] Endpoint reachability: reachable → not serving MCP
- [security regression] Stability: fail → unverified
- [security regression] HTTPS: fail → unverified
- [security regression] Authorization: partial → unverified
- [security regression] Tool safety: pass → unverified
- [security regression] Transport: pass → fail
- [functional regression] Capabilities: fail → unverified
- [functional regression] Tool coverage: 100 → unverified
- [functional] First check of Schema quality: unverified

### 2026-10-26 (score 61, 0)

- [security] Tool “audit_mcp_server_manifest” rewrote its description, which is the text the model reads
- [security] Tool “audit_outbound_egress_url” rewrote its description, which is the text the model reads
- [security] Tool “audit_response_factuality” rewrote its description, which is the text the model reads
- [security] Tool “detect_agent_action_loop” rewrote its description, which is the text the model reads
- [security] Tool “detect_mcp_schema_rugpull” rewrote its description, which is the text the model reads
- [security] Tool “execute_python_sandbox” rewrote its description, which is the text the model reads
- [security] Tool “redact_sensitive_pii” rewrote its description, which is the text the model reads
- [security] Tool “sanitize_tool_output_stream” rewrote its description, which is the text the model reads
- [security] Tool “validate_mcp_sampling_request” rewrote its description, which is the text the model reads
- [security] Tool “validate_sql_query_ast” rewrote its description, which is the text the model reads
- [security] Tool “verify_a2a_agent_card” rewrote its description, which is the text the model reads
- [security] Tool “verify_agent_command_safety” rewrote its description, which is the text the model reads

### 2026-10-21 (score 61, +4)

- [security regression] Stability: unverified → fail
- [security regression] A breaking change shipped without a version bump: still 2.5.0
- [security regression] Tool “agent_loop_circuit_breaker” was removed
- [security regression] Tool “agent_mcp_manifest_auditor” was removed
- [security regression] Tool “agent_sql_guard” was removed
- [security regression] Tool “code_sandbox” was removed
- [security regression] Tool “guardrail_verify” was removed
- [security regression] Tool “mcp_rugpull_diff” was removed
- [security regression] Tool “mcp_sampling_guard” was removed
- [security regression] Tool “scrub_pii” was removed
- [security regression] Tool “tool_output_firewall” was removed
- [functional regression] Schema quality: pass → fail
- [functional improvement] Tool coverage: 27% → 100%
- [functional] First check of Tool coverage: 100
- [functional] Schema quality: fair → excellent
- [functional] Destructive annotations: pass → 0
- [functional] New tool “audit_mcp_server_manifest”
- [functional] New tool “audit_outbound_egress_url”
- [functional] New tool “audit_response_factuality”
- [functional] New tool “detect_agent_action_loop”
- [functional] New tool “detect_mcp_schema_rugpull”
- [functional] New tool “execute_python_sandbox”
- [functional] New tool “redact_sensitive_pii”
- [functional] New tool “sanitize_tool_output_stream”
- [functional] New tool “validate_mcp_sampling_request”
- [functional] New tool “validate_sql_query_ast”
- [functional] New tool “verify_a2a_agent_card”
- [functional] New tool “verify_agent_command_safety”

### 2026-10-20 (score 57)

First indexed and scored.

## MCP tools (12)

### `verify_agent_command_safety` (~168 tokens)

Autonomous Agent Pre-Execution Command & Prompt Injection Gate

Evaluates untrusted agent prompts, bash commands, and tool arguments to block prompt injection, destructive shell commands, and private-key exfiltration. Use when screening a general shell command or user prompt before autonomous agent execution; for statically validating SQL database queries at the AST level, use validate_sql_query_ast instead. Stateless read-only edge call (max 32 KB, 60 RPM): accepts 'payload' ('action_type' defaults to 'bash_command'), supports '?demo=true' or L402 (20 sats), and returns HTTP 400 if invalid.

Input parameters:

- `action_type` (string): Execution context of the payload; defaults to 'bash_command' when omitted.
- `payload` (string, required): Raw command string, script snippet, or untrusted prompt to inspect for safety violations.

Output parameters:

- `risk_score` (number): Normalized threat probability from 0.0 (benign) to 1.0 (critical).
- `safe` (boolean): True when zero injection, exfiltration, or destructive patterns are found.
- `threats_detected` (array): List of specific threat categories triggered by the payload.
- `verdict` (string): Enforcement decision ('ALLOW', 'BLOCK', or 'QUARANTINE').

### `validate_sql_query_ast` (~190 tokens)

AST-Level Read-Only SQL Query & Injection Safety Validator

Performs static AST syntax tree parsing on SQL statements to verify strictly read-only SELECT clauses and reject mutations, stacked queries, or comment bypasses. Use when verifying that an agent-authored SQL statement is strictly read-only before running it on a database; for screening non-SQL bash commands or general agent prompts, use verify_agent_command_safety instead. Stateless read-only edge call (max 32 KB, 60 RPM): accepts 'sql_statement' ('dialect' defaults to 'postgres'), supports '?demo=true' or L402 (15 sats), and returns HTTP 400 if invalid.

Input parameters:

- `dialect` (string): SQL dialect ruleset used for AST token inspection; defaults to 'postgres' when omitted.
- `sql_statement` (string, required): SQL statement string to statically parse and audit for read-only SELECT compliance (e.g. 'SELECT id, email FROM users WHERE active = 1').

Output parameters:

- `blocked_clauses` (array): List of destructive or injection SQL tokens detected.
- `safe_read_only` (boolean): True when the query is a pure SELECT statement with zero injection clauses.
- `tables_accessed` (array): Table identifiers referenced in FROM and JOIN clauses.
- `verdict` (string): Execution decision ('ALLOW_EXECUTION' or 'BLOCK_EXECUTION').

### `execute_python_sandbox` (~183 tokens)

AST-Audited Ephemeral Python Code Execution Sandbox

Audits Python source code for forbidden syscalls (os, subprocess, socket, open, eval) inside an AST-governed WebAssembly sandbox with no host filesystem access. Use when executing LLM-generated Python snippets safely; for checking shell commands without executing code, use verify_agent_command_safety instead. Stateless read-only edge call (max 32 KB, 60 RPM): accepts 'python_script' ('timeout_profile' defaults to 'fast_1s'), supports '?demo=true' or MPP (250 sats), and returns HTTP 400 if invalid.

Input parameters:

- `python_script` (string, required): Python 3 source script to audit and execute strictly inside an isolated AST-governed, non-privileged WebAssembly sandbox with no host filesystem access.
- `timeout_profile` (string): Execution resource ceiling; defaults to 'standard_5s' when omitted.

Output parameters:

- `ast_security_audit` (object): Static syscall audit summary including flagged_calls and safety_verdict.
- `exit_code` (integer): Process exit status (0 on clean execution, 1 on security block).
- `simulated_stdout` (string): Captured standard output from the isolated execution.
- `syntax_valid` (boolean): True if the Python source parses without syntax errors.

### `audit_mcp_server_manifest` (~186 tokens)

External MCP Server Manifest & Tool-Poisoning Security Auditor

Scans an external MCP server's initial tools/list manifest to detect tool-name shadowing (read_file, bash) and hidden prompt-injection directives in descriptions. Use when vetting a newly discovered third-party MCP server before mounting it into an agent session; for comparing a previously pinned MCP baseline against a live mid-session tools/list response, use detect_mcp_schema_rugpull instead. Stateless read-only edge call (max 32 KB, 60 RPM): accepts 'manifest_json' ('audit_profile' defaults to 'strict'), supports '?demo=true' or L402 (25 sats), and returns HTTP 400 if invalid.

Input parameters:

- `audit_profile` (string): Sensitivity level for flagging system-tool name shadowing; defaults to 'strict' when omitted.
- `manifest_json` (string, required): JSON string or object containing the external MCP server's tools array to audit.

Output parameters:

- `manifest_safe` (boolean): True if zero shadowed tools or description injections are found.
- `security_grade` (string): Overall manifest safety grade ('A+', 'C', or 'F').
- `shadowed_tools` (array): Tool names attempting to shadow host privileges (e.g. 'bash', 'read_file').
- `suspicious_tools` (array): Tools containing covert prompt-injection or exfiltration instructions.

### `detect_mcp_schema_rugpull` (~209 tokens)

Mid-Session MCP tools/list Rug-Pull & Schema Mutation Detector

Computes structural diffs between a pinned baseline MCP tools/list snapshot and a live server response to catch mid-session description or schema tampering. Use when verifying that a connected MCP server has not mutated its tool definitions mid-session; for performing a first-time static security audit on a single MCP manifest, use audit_mcp_server_manifest instead. Stateless read-only edge call (max 32 KB, 60 RPM): accepts 'baseline_tools_json' ('current_tools_json' defaults to 'standard'), supports '?demo=true' or L402 (20 sats), and returns HTTP 400 if invalid.

Input parameters:

- `baseline_tools_json` (string, required): JSON-serialized baseline tools/list array captured when the MCP server was first trusted.
- `current_tools_json` (string): JSON-serialized live tools/list array to compare against the baseline; defaults to '[]' when omitted.
- `diff_mode` (string): Comparison strictness across descriptions and inputSchema; defaults to 'full_structural' when omitted.

Output parameters:

- `mutated_tools` (array): Detailed list of modified tools and their exact mutation type.
- `rugpull_detected` (boolean): True if any tool was surreptitiously added or had its description/schema altered.
- `timestamp_utc` (string): ISO-8601 UTC timestamp of the diff audit.
- `verdict` (string): Connection verdict ('SCHEMA_INTEGRITY_VERIFIED' or 'REVOKE_MCP_SERVER_IMMEDIATELY').

### `validate_mcp_sampling_request` (~179 tokens)

Inbound MCP sampling/createMessage Reverse-Injection Firewall

Audits inbound MCP sampling/createMessage callbacks from connected servers to block cross-server context snooping (includeContext='allServers') and token quota hijacking. Use when filtering a connected MCP server's request to sample the host LLM; for scrubbing text returned from a standard synchronous tools/call response, use sanitize_tool_output_stream instead. Stateless read-only edge call (max 32 KB, 60 RPM): accepts 'sampling_request_json' ('max_tokens_ceiling' defaults to '512'), supports '?demo=true' or L402 (20 sats), and returns HTTP 400 if invalid.

Input parameters:

- `max_tokens_ceiling` (string): Maximum completion token quota allowed for server callbacks; defaults to '2048' when omitted.
- `sampling_request_json` (string, required): JSON string or object representing the inbound MCP sampling/createMessage payload.

Output parameters:

- `recommended_action` (string): Host enforcement action ('APPROVE_SAMPLING_REQUEST' or 'REJECT_SAMPLING_REQUEST').
- `sampling_safe` (boolean): True if the sampling request respects context isolation and token caps.
- `timestamp_utc` (string): ISO-8601 UTC timestamp of the inspection.
- `violations_detected` (array): Specific policy violations found in the sampling request.

### `sanitize_tool_output_stream` (~186 tokens)

Indirect Prompt-Injection, Invisible Unicode & Beacon Output Scrubber

Sanitizes untrusted external tool or web-scrape outputs by stripping invisible Unicode Tag characters (U+E0000), neutralizing Markdown tracking image beacons, and defusing indirect prompt overrides. Use when cleaning raw third-party tool outputs or scraped web text before inserting it into the LLM context window; for redacting personal identifiable information (emails, SSNs, keys) from outbound text, use redact_sensitive_pii instead. Read-only edge call (32 KB, 60 RPM): accepts 'raw_tool_output' (with enum defaults), uses '?demo=true' or L402 (15 sats); HTTP 400 if invalid.

Input parameters:

- `raw_tool_output` (string, required): Untrusted text returned by an external tool, webhook, or web scraper.
- `wrap_tag` (string): XML isolation wrapper applied around the sanitized text; defaults to 'tool_data' when omitted.

Output parameters:

- `injection_risk_score` (number): Threat score from 0.0 to 0.99 prior to sanitization.
- `safe_for_context_ingestion` (boolean): True once the sanitized payload is safe to feed to the LLM.
- `sanitized_output` (string): Cleaned text with invisible Unicode, tracking beacons, and overrides neutralized.
- `threats_neutralized` (array): List of smuggling or injection vectors stripped from the input.

### `redact_sensitive_pii` (~169 tokens)

Zero-PII & Secret Credential Redaction Engine

Redacts emails, phone numbers, SSNs, credit cards, API keys, and crypto private keys from text before logging or external transmission. Use when stripping personal data or secret credentials from prompts, logs, or outbound messages; for neutralizing indirect prompt injections and invisible Unicode in inbound tool responses, use sanitize_tool_output_stream instead. Stateless read-only edge call (max 32 KB, 60 RPM): accepts 'text' ('redaction_style' defaults to 'typed_token'), supports '?demo=true' or L402 (25 sats), and returns HTTP 400 if invalid.

Input parameters:

- `redaction_style` (string): Placeholder format used for redacted spans; defaults to 'typed_token' when omitted.
- `text` (string, required): Input text containing potential PII or secret keys to redact.

Output parameters:

- `categories_redacted` (array): Types of sensitive data detected (e.g. 'EMAIL', 'SSN', 'PRIVATE_KEY').
- `redactions_count` (integer): Total number of sensitive entities redacted.
- `scrubbed_text` (string): Sanitized text with all sensitive entities replaced by redaction tokens.
- `zero_pii_certified` (boolean): True when post-scan verification confirms zero residual PII patterns.

### `detect_agent_action_loop` (~187 tokens)

Autonomous Agent Action-Trace Loop & Token-Burn Circuit Breaker

Analyzes an agent's recent tool-call trace to detect identical retry spins, A/B state oscillation, and runaway token burn. Use when monitoring autonomous agent execution loops to decide whether to continue or trip a token-burn breaker; for inspecting a single command payload for security threats before execution, use verify_agent_command_safety instead. Stateless read-only edge call (max 32 KB, 60 RPM): accepts 'action_trace_json' ('sensitivity' defaults to 'strict_2_repeats'), supports '?demo=true' or L402 (15 sats), and returns HTTP 400 if invalid.

Input parameters:

- `action_trace_json` (string, required): JSON-serialized array of recent agent actions (e.g. '[{"tool":"search","status":"err"}]').
- `sensitivity` (string): Cycle detection threshold; defaults to 'standard_3_repeats' when omitted.

Output parameters:

- `circuit_breaker_action` (string): Recommended control action ('CONTINUE_EXECUTION' or 'HALT_AND_ESCALATE_OR_SWITCH_TOOL').
- `estimated_tokens_saved` (integer): Projected token burn prevented by tripping the circuit breaker.
- `loop_detected` (boolean): True if a repetitive retry spin or A/B oscillation cycle is active.
- `loop_pattern` (string): Detected cycle topology ('NONE', 'IDENTICAL_RETRY_SPIN', or 'ABAB_STATE_OSCILLATION').

### `verify_a2a_agent_card` (~200 tokens)

A2A /.well-known/agent.json Identity & Capability Card Verifier

Verifies a remote agent's A2A /.well-known/agent.json discovery card, checking JWS/DID signature binding, endpoint HTTPS enforcement, and capability spoofing. Use when authenticating a peer autonomous agent before delegating subtasks or exchanging payments over A2A; for auditing an MCP tool server's tools/list manifest rather than a peer agent's A2A card, use audit_mcp_server_manifest instead. Read-only edge call (32 KB, 60 RPM): accepts 'agent_card_json' ('trust_policy' defaults to 'require_https_and_did'), supports '?demo=true' or L402 (15 sats), and returns HTTP 400 if invalid.

Input parameters:

- `agent_card_json` (string, required): JSON string or object of the peer agent's /.well-known/agent.json descriptor.
- `trust_policy` (string): Verification policy for DID/DNS binding; defaults to 'require_https_and_did' when omitted.

Output parameters:

- `agent_id` (string): Canonical identifier of the verified peer agent.
- `card_valid` (boolean): True when the A2A agent card schema, transport, and identity claims pass verification.
- `delegation_verdict` (string): Handshake decision ('SAFE_TO_DELEGATE_A2A' or 'REJECT_UNVERIFIED_AGENT').
- `spoofing_flags` (array): List of impersonation or downgrade anomalies detected.

### `audit_outbound_egress_url` (~198 tokens)

Outbound Agent URL SSRF, Metadata & Secret Exfiltration Firewall

Static validation firewall for outbound network destination URIs against cloud metadata IP addresses (169.254.169.254), RFC 1918 private IP blocks, and query-string secret exfiltration before agent dispatch. Use when validating a user-supplied or tool-supplied URI destination before network dispatch; for auditing local shell commands or general prompts, use verify_agent_command_safety instead. Read-only edge call (32 KB, 60 RPM): accepts 'destination_uri' ('egress_policy' defaults to 'block_private_and_metadata'), supports '?demo=true' or L402 (15 sats), and returns HTTP 400 if invalid.

Input parameters:

- `destination_uri` (string, required): Outbound HTTP/HTTPS destination URI to inspect before fetching (e.g. 'https://api.example.com/v1/webhook').
- `egress_policy` (string): Network egress policy; defaults to 'block_private_and_metadata' when omitted.

Output parameters:

- `egress_allowed` (boolean): True if the URL targets a safe public endpoint with no exfiltration tokens.
- `exfiltration_query_params` (array): Query parameters suspected of carrying leaked secrets or base64 payloads.
- `ssrf_risk_detected` (boolean): True if loopback, RFC1918 private IP, or cloud metadata IP is targeted.
- `verdict` (string): Firewall decision ('ALLOW_EGRESS_REQUEST' or 'BLOCK_SSRF_OR_EXFIL').

### `audit_response_factuality` (~197 tokens)

Draft Agent Response Groundedness & Hallucination Auditor

Cross-references factual claims, numbers, and citations in a draft agent response against retrieved source passages to flag unsupported hallucinations before delivery. Use when verifying that a generated answer is 100% grounded in source evidence before returning it to the user; for scrubbing inbound tool outputs for prompt-injection and invisible Unicode, use sanitize_tool_output_stream instead. Stateless read-only edge call (max 32 KB, 60 RPM): accepts 'claim' ('context' defaults to 'standard'), supports '?demo=true' or L402 (25 sats), and returns HTTP 400 if invalid.

Input parameters:

- `claim` (string, required): Draft agent response or factual statement to verify for hallucinations.
- `context` (string, required): Authoritative source text or retrieved RAG passages that the claim must be grounded in.
- `strictness` (string): Entailment threshold required for numbers and dates; defaults to 'exact_entailment' when omitted.

Output parameters:

- `groundedness_score` (number): Factual support ratio from 0.0 (fabricated) to 1.0 (fully entailed).
- `unverified_claims` (array): Hallucinated or unsupported assertions flagged for removal.
- `verdict` (string): Audit outcome ('PASS', 'SUSPICIOUS', or 'FAIL').
- `verified_claims` (array): Assertions directly entailed by the provided source context.

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/stand-out-of-my-sun-diogenes-agent-guardrails-mcp/mcp-agent-guardrails#diagnostics

## Score history

- 2026-11-05: 65
- 2026-11-04: 65
- 2026-11-03: 65
- 2026-11-02: 18
- 2026-11-01: 61
- 2026-10-31: 61
- 2026-10-30: 61
- 2026-10-29: 61
- 2026-10-28: 61
- 2026-10-27: 61
- 2026-10-26: 61
- 2026-10-25: 61
- 2026-10-24: 61
- 2026-10-23: 61
- 2026-10-22: 61
- 2026-10-21: 61
- 2026-10-20: 57

## Common questions

### What is the Diogenes Agent Guardrails & Sandbox MCP server?

Diogenes Agent Guardrails & Sandbox MCP is listed in the public MCP registry as io.github.stand-out-of-my-sun/diogenes-agent-guardrails-mcp. Autonomous agent guardrails, bash sandboxing, MCP rugpull diffing & prompt firewalls. This page covers its hosted endpoint (https://diogenes-micro-gateway.corinth.workers.dev/mcp/agent-guardrails).

### Is the Diogenes Agent Guardrails & Sandbox MCP server safe to use?

Diogenes Agent Guardrails & Sandbox MCP scores 65 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the Diogenes Agent Guardrails & Sandbox MCP server expose?

Diogenes Agent Guardrails & Sandbox MCP exposes 12 tools: verify_agent_command_safety, validate_sql_query_ast, execute_python_sandbox, audit_mcp_server_manifest, detect_mcp_schema_rugpull, and 7 more. Their descriptions and schemas cost roughly 2,252 tokens of context every time the server is loaded.

### Does the Diogenes Agent Guardrails & Sandbox MCP server require authentication?

No. We connected to Diogenes Agent Guardrails & Sandbox MCP without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the Diogenes Agent Guardrails & Sandbox MCP server still maintained?

Diogenes Agent Guardrails & Sandbox MCP is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

## Links

- Remote endpoint: https://diogenes-micro-gateway.corinth.workers.dev/mcp/agent-guardrails
- Repository: https://github.com/stand-out-of-my-sun/diogenes-agent-guardrails-mcp
- Website: https://diogenes-micro-gateway.corinth.workers.dev/
- Changelog RSS feed: https://verifymcp.io/servers/stand-out-of-my-sun-diogenes-agent-guardrails-mcp/mcp-agent-guardrails.xml
- Changelog JSON feed: https://verifymcp.io/servers/stand-out-of-my-sun-diogenes-agent-guardrails-mcp/mcp-agent-guardrails.json
- HTML version of this page: https://verifymcp.io/servers/stand-out-of-my-sun-diogenes-agent-guardrails-mcp/mcp-agent-guardrails