AI agents use requirement_manager to create or update resources in AI Develop Assistant — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your AI Develop Assistant environment.
The tool's core functions (classify, deduplicate, validate requirements) constitute data modification and organization activities. While it processes development artifacts rather than executing code or deleting data, it clearly writes/modifies the requirement information state.
From the tool's definition Tool description indicates it manages requirement documents through 'classification', 'deduplication', and 'validation' of requirement information. These are reversible operations that modify and organize data (create, update, reorganize requirements).
Documented attack patterns abuse exactly the kind of access requirement_manager gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and AI Develop Assistant, and nothing reaches the server without passing your rules. This is the rule we recommend for requirement_manager:
{
"version": "1",
"default": "deny",
"tools": {
"requirement_manager": {
"limits": [
{
"counter": "requirement_manager_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}requirement_manager stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
智能需求文档管理器 - 智能分类、去重、验证需求信息. It is categorised as a Write tool in the AI Develop Assistant MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the AI Develop Assistant MCP server in PolicyLayer and add a rule for requirement_manager: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches AI Develop Assistant. Nothing to install.
requirement_manager is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the requirement_manager rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for requirement_manager. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
requirement_manager is provided by the AI Develop Assistant MCP server (jiemobasixiangcai/ai-develop-assistant). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from AI Develop Assistant, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
5 AI Develop Assistant tools catalogued and risk-classified — across an index of 43,000+ MCP servers.