High-risk tools in VibeServe
26 of the 126 tools in VibeServe are classified as high risk. This page profiles those tools specifically, with recommended policy actions and the attack patterns that target them.
Every operation listed below is an action PolicyLayer recommends controlling at the transport layer. Open any tool to see the full profile, risk score, and YAML policy snippet.
Tools at high risk
-
browser_automationExecuteControl a headless browser using a Playwright-based DSL. Use for automation, scraping, or web interactions. Isolated incognito context is created per call.
-
codegraph_buildExecuteBuild a knowledge graph from an indexed repo — creates call graph, import resolution, class hierarchy, and communities. Prerequisite before using other codegraph_* tools.
-
deterministic_brainExecuteQuery the deterministic brain neuro-symbolic AI reasoning engine with 5 lanes: coding, business_logic, agent_brain, tool_calling, cross_domain. Auto-routes by default.
-
google_cloud_run_queryExecuteRun a Google Cloud SQL query
-
governance_kill_switchExecuteKill switch — disable all Big-Homie agent operations
-
llm_completeExecuteComplete a prompt through Big-Homie
-
nanobot_run_skillExecuteExecute a nanobot skill by name
-
nanobot_schedule_cronExecuteRegister a cron job in nanobot
-
nanobot_sendExecuteSend a message through any nanobot channel
-
run_biomeExecuteRun Biome linter/formatter
-
run_buildExecuteRun production build
-
run_installExecuteRun package installation
-
run_npm_auditExecuteRun npm audit for dependency security
-
run_playwrightExecuteRun Playwright E2E tests
-
run_semgrepExecuteRun Semgrep SAST scan on the project
-
run_tscExecuteRun TypeScript compiler check
-
supabase_queryExecuteQuery a Supabase table.
-
trigger_pipelineExecuteTrigger a Nexus Alpha build pipeline for a list of repositories.
-
twilio_send_smsExecuteSend an SMS via Twilio
-
vibe_benchmarkExecuteRun a benchmarking loop with ASCII graphs.
-
vibe_build_proExecuteFull professional build: upgrade design -> architect -> code -> verify.
-
vibe_checkExecuteRun all VibeCoder quality gates (type safety, lint, security, bundle, deps, structure) and return a scored build report with letter grade (S/A/B/C/D/F). Perfect for vibe coders ...
-
vibe_deployExecuteGenerate deployment configs for Vercel, Docker, static hosting.
-
vibe_iterateExecuteContinuous improvement loop: critique -> repair -> verify -> repeat.
-
vs_ecc_agent_shieldExecuteRun ECC AgentShield security scan on provided code content. Checks for secrets, permission risks, hook injection, MCP risks, and config issues.
-
vs_opencode_executeExecuteExecute a coding task using the OpenCode CLI agent.
Attacks that target this class
High-risk tools in any server share these documented attack patterns. Each links to the full case and the defensive policy.