Skip to main content
Glama

Start a worker session

agent_start

Start a new reverse-engineering session: configure a worker model, disable confirmations, optionally navigate to a target URL, and send the initial task, returning a session ID for subsequent agent polling and reading conclusions.

Instructions

开一个新的逆向会话:配置 worker 模型(默认沿用浏览器里已配的)+关掉工具确认门、(可选)导航到目标 URL、建工作目录、用便宜的 worker 模型以【AI 辅助模式】发出第一轮(task 作为第 0 条 user)。返回 tid,后续都用它。推荐流程:GPT/CLI 委派模式下,agent_start 返回后继续 agent_poll/agent_wait_for_stop 观察阶段状态,settled 后用 agent_read_brief/agent_read 读取结论。绝不读/写 worker 的 key——key 由用户预先在浏览器里配好;这里只校验 hasKey=true。worker 选标准/快档别选推理档(推理档在长工具循环里易漂移成纯文字),如 deepseek-v4-flash / qwen-turbo / glm。

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
taskYes逆向目标(成为 convo[0] 的 user 内容),例如:还原 xxx.com 的 X-S 签名并 Node 实打接口
modelNoworker 模型名(省略=沿用浏览器配置)。flash 档,如 deepseek-v4-flash(中转)/deepseek-chat(直连)
assistNo默认 true=阶段门停。false=全自动
providerNo切到这个 provider 再跑(如 deepseek/zhipu/custom),省略=沿用浏览器当前 active。不碰 key,只校验该 provider 的 hasKey
maxRoundsNo单轮最大工具回合,默认 80
targetUrlNo开跑前把当前 tab 导航到这个 URL
workspaceRootNo工作目录绝对路径(省略=在 FRX_WORKSPACE_ROOT 下自动建一个)
ensureConfirmOffNo默认 true=关掉工具确认门(无人值守必需)
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description fully carries the burden. It details configuration steps (worker model, confirm gate, navigation, directory creation), return value (tid), key handling (only check hasKey), and mode behavior (assist vs auto). Slightly lacks explicit blocking/non-blocking behavior, but overall transparent.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is lengthy but well-structured: starts with core purpose, then workflow, then warnings, then recommendations. Every sentence is informative; minimal redundancy. Could be slightly more concise, but highly effective.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given 8 parameters, no output schema, and no annotations, the description covers all major aspects: purpose, workflow, behavioral details, parameter semantics, and caveats. It provides sufficient context for correct tool invocation without needing external references.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% (all 8 parameters described in schema). Description adds value beyond schema: clarifies 'assist' default and its meaning, gives model examples, explains 'targetUrl' effect, and describes default for 'workspaceRoot'. Exceeds baseline 3.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states it starts a new reverse session ('开一个新的逆向会话'), identifies the specific resource (session) and action (start), and is distinct from sibling tools like agent_poll (observation) and agent_read (reading conclusions).

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly recommends a workflow: after agent_start, use agent_poll/agent_wait_for_stop, then agent_read_brief/agent_read. Also advises against reading/writing worker keys and suggests model selection (standard/fast not reasoning). Provides clear when-to-use and when-not-to-use guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/WhiteNightShadow/frx-director-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server