AI agents use init_qa_knowledge to create or update resources in Mk Qa Master — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Mk Qa Master environment.
This tool creates a new file (qa-knowledge.md) at the project root. It is a file creation/write operation — reversible by deleting the file — so it falls under Write. No code is executed, no data is destroyed, and no financial transaction occurs. Severity is medium because writing files to a project root could overwrite existing content or introduce unexpected files into a repository.
From the tool's definition 建立 qa-knowledge.md 起手範本 (creates a qa-knowledge.md template file at the project root PROJECT_ROOT)
Documented attack patterns abuse exactly the kind of access init_qa_knowledge gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Mk Qa Master, and nothing reaches the server without passing your rules. This is the rule we recommend for init_qa_knowledge:
{
"version": "1",
"default": "deny",
"tools": {
"init_qa_knowledge": {
"limits": [
{
"counter": "init_qa_knowledge_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}init_qa_knowledge 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.
在受測專案根 (PROJECT_ROOT) 建立 qa-knowledge.md 起手範本，. It is categorised as a Write tool in the Mk Qa Master MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Mk Qa Master MCP server in PolicyLayer and add a rule for init_qa_knowledge: 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 Mk Qa Master. Nothing to install.
init_qa_knowledge 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 init_qa_knowledge 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 init_qa_knowledge. 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.
init_qa_knowledge is provided by the Mk Qa Master MCP server (kao273183/mk-qa-master). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Mk Qa Master, 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.
22 Mk Qa Master tools catalogued and risk-classified — across an index of 43,000+ MCP servers.