Read or manage the podcli knowledge base. These are .md files that provide context about the podcast (hosts, style, audience, etc). Always read the knowledge base before suggesting or creating clips.
AI agents use knowledge_base to create or update resources in Podcli — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Podcli environment.
The tool description explicitly states it can both read and manage knowledge base files. Since it spans Read and Write categories, the most severe applicable category is Write. The ability to modify context files (hosts, style, audience info) used to guide clip creation could have meaningful downstream effects if misused, warranting medium severity.
From the tool's definition 'Read or manage the podcli knowledge base' and 'These are .md files' — the tool both reads and manages (writes/modifies) markdown files
Documented attack patterns abuse exactly the kind of access knowledge_base gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Podcli, and nothing reaches the server without passing your rules. This is the rule we recommend for knowledge_base:
{
"version": "1",
"default": "deny",
"tools": {
"knowledge_base": {
"limits": [
{
"counter": "knowledge_base_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}knowledge_base 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.
Read or manage the podcli knowledge base. These are .md files that provide context about the podcast (hosts, style, audience, etc). Always read the knowledge base before suggesting or creating clips. It is categorised as a Write tool in the Podcli MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Podcli MCP server in PolicyLayer and add a rule for knowledge_base: 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 Podcli. Nothing to install.
knowledge_base 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 knowledge_base 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 knowledge_base. 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.
knowledge_base is provided by the Podcli MCP server (nmbrthirteen/podcli). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Podcli, 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.
17 Podcli tools catalogued and risk-classified — across an index of 43,000+ MCP servers.