This section is a quick, utilitarian map of the tool surface. For the complete operational contract (concepts, workflows, failure modes, and full stdlib reference), ask you AI to call kahlo_mcp_about to tell you more!
kahlo_devices_list: list connected devices ready for instrumentation.
kahlo_devices_get (device_id): detailed device info (model/version/transport) and readiness signals.
kahlo_devices_health (device_id): health check for ADB + frida-server presence/running.
kahlo_processes_list (device_id, scope?): list running processes (pid + name). Use this before mode="attach" to get the exact process name.
kahlo_adb_command (command[], device_id?, timeout_ms?): run an ADB command with the configured ADB binary. For root-only paths, wrap with su -c (e.g., ['shell','su','-c','ls /data/data/<pkg>']).
Targets (attach/spawn + lifecycle)
kahlo_targets_ensure:
Purpose: create/ensure a target (an instrumented app process hosting the in-process orchestrator).
mode="attach": package is the process name from kahlo_processes_list (often not the Android package id).
mode="spawn": package is the Android package id (e.g., com.example.app).
gating="spawn" requires bootstrap (early hooks while app is suspended, then it resumes automatically).
gating="child" is present but does not work on Android (zygote forks child processes). For multi-process apps, discover com.example.app:process names and attach separately.
kahlo_targets_status (target_id): target lifecycle + orchestrator agent state (use to diagnose crashes/reinjecting).
kahlo_targets_detach (target_id): detach cleanly; cancels all jobs for that target.
Jobs (run instrumentation code)
kahlo_jobs_start:
Purpose: start an isolated job (one Frida script instance) inside a target.
Params: target_id, type? (oneshot|interactive|daemon), ttl?, params?, and module:
kahlo_modules_promoteDraft (draft_id, name, version_strategy, notes?): freeze a draft into a versioned module (patch|minor|major).
kahlo_modules_promoteFromJob (job_id, name, version_strategy, notes?): promote a tested job directly into a versioned module.
Stdlib (ctx.stdlib)
Every job script gets a preloaded standard library under ctx.stdlib. This is the preferred way to do common "instrumentation chores" (stack capture, safe Java calls, overload-safe hooks, bytes/strings, Intent parsing, etc.) without rewriting boilerplate.
Source of truth: kahlo-mcp/src/backend/jobs/jobScriptStdlib.js.
Top-level helpers
ctx.stdlib.isJavaAvailable(): returns true if Java APIs can be used in this process.
ctx.stdlib.getJavaBridge(): returns the Java bridge handle (or null).
ctx.stdlib.requireJava(operation?): returns the Java bridge or throws with a helpful message.
Ainda não há lista de ferramentas. Servidores remotos recebem uma após a primeira verificação de saúde.
Perguntas frequentes
Perguntas
Como instalo Kahlo?
Copie nesta página a configuração stdio para Cursor, Claude Desktop, Claude Code ou VS Code. Ela usa a instalação do publicador, não um proxy do PlugMCP.
Kahlo precisa de uma chave de API?
Este registro não lista nenhuma variável de ambiente obrigatória.
Qual transporte Kahlo usa?
Kahlo roda localmente via stdio.
Quando Kahlo foi verificado pela última vez?
Última verificação: 2026-10-27. A Frida MCP server to enable autonomous AI assistance for Android instrumentation