// SCAN REPORT

Vibecheck

103 tools. 30 can modify or destroy data without limits.

3 destructive tools with no built-in limits. Policy required.

Last updated: 6 Jun 2026

30 can modify or destroy data

73 read-only

103 tools total

Free to start. No card required.

TOOL MAP

Read (73) Write / Execute (27) Destructive / Financial (3)

WHAT CAN GO WRONG

Destructive tools (forge_reset, intent_clear, middleware_clear_traces) permanently delete resources. There is no undo. An agent calling these in a retry loop causes irreversible damage.

Write operations (context_update_embeddings, docguard_config, firewall_lockdown) modify state. Without rate limits, an agent can make hundreds of changes in seconds — faster than any human can review or revert.

Execute tools (agent_graph_run, agent_tool_execute, hook_pre_generation) trigger processes with side effects. Builds, notifications, workflows — all fired without throttling.

RECOMMENDED RULES

Deny destructive operations

{
  "forge_reset": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations

{
  "context_update_embeddings": {
    "limits": [
      {
        "counter": "context_update_embeddings_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "agent_gate_history": {
    "limits": [
      {
        "counter": "agent_gate_history_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

ALL 103 VIBECHECK TOOLS

DESTRUCTIVE 3 tools

Destructive forge_reset Destructive intent_clear Destructive middleware_clear_traces

EXECUTE 5 tools

Execute agent_graph_run Execute agent_tool_execute Execute hook_pre_generation Execute isl_from_task Execute prompt_build

READ 73 tools

Read agent_gate_history Read agent_intent_declare Read agent_runtime_init Read agent_tools_list Read agent_transcript_get Read context_conventions Read context_for_file Read context_gather Read context_structure Read docguard_evaluate Read docguard_find_canonical Read docguard_quick_check Read docguard_scan Read docguard_stats Read docguard_suggest_location Read excessive-claims Read finding_explanation Read finding_types_list Read firewall_evaluate Read firewall_extract_claims Read firewall_get_mode Read firewall_get_tier Read firewall_quick_check Read firewall_should_block Read firewall_status Read firewall_tier_stats Read forge_analyze Read forge_contract Read forge_memory_query Read forge_memory_record Read forge_memory_stats Read forge_memory_timeline Read forge_phase Read forge_rule_feedback Read forge_rules Read forge_status Read ghost-env Read ghost-file Read ghost-route Read glossary_list Read glossary_lookup Read glossary_search Read intent_check Read intent_declare Read intent_extend Read intent_get Read intent_history Read intent_restrict Read isl_answer_questions Read isl_print Read isl_repair_ast Read isl_translate Read isl_validate_ast Read low-confidence Read middleware_intent_history Read middleware_traces Read onboarding_status Read prompt_get_task Read prompt_quick_check Read prompt_templates Read prompt_verify Read quota_info Read quota_status Read truthpack_env Read truthpack_query Read truthpack_routes Read truthpack_validate Read validation_code Read validation_detect_drift Read validation_detect_hallucinations Read validation_verify Read validation_verify_imports Read vibecoder_services

FAQ

Can an AI agent delete data through the Vibecheck MCP server? +

Yes. The Vibecheck server exposes 3 destructive tools including forge_reset, intent_clear, middleware_clear_traces. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through Vibecheck? +

The Vibecheck server has 22 write tools including context_update_embeddings, docguard_config, firewall_lockdown. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Vibecheck.

How many tools does the Vibecheck MCP server expose? +

103 tools across 4 categories: Destructive, Execute, Read, Write. 73 are read-only. 30 can modify, create, or delete data.

How do I enforce a policy on Vibecheck? +

Register the Vibecheck MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

// RELATED SERVERS

Other MCP servers with similar tools.

Starter policies for each. Same risk classification, live on your fleet in minutes.

Enforce policy on every Vibecheck tool call.

Deterministic rules across all 103 Vibecheck tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

SECURE VIBECHECK →

Free to start. No card required.

4,600+ MCP servers and 31,000+ tools scanned and risk-classified.

Vibecheck

// TOOL MAP

// WHAT CAN GO WRONG

// RECOMMENDED RULES

// ALL 103 VIBECHECK TOOLS

// FAQ