AI agents call peek_agent_notes to retrieve information from Claude Slack without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves agent notes, which is a read operation with no side effects. However, the 'medium' severity reflects that accessing another agent's notes could expose sensitive project information, collaboration details, or internal reasoning that an unauthorized agent should not see—creating a potential confidentiality issue.
From the tool's definition Tool name 'peek_agent_notes' and description 'Peek at another agent' indicate retrieval of data (agent notes) without modification. The verb 'peek' suggests viewing/reading access.
Documented attack patterns abuse exactly the kind of access peek_agent_notes gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Claude Slack, and nothing reaches the server without passing your rules. This is the rule we recommend for peek_agent_notes:
{
"version": "1",
"default": "deny",
"tools": {
"peek_agent_notes": {}
}
} peek_agent_notes is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Peek at another agent. It is categorised as a Read tool in the Claude Slack MCP Server, which means it retrieves data without modifying state.
Register the Claude Slack MCP server in PolicyLayer and add a rule for peek_agent_notes: 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 Claude Slack. Nothing to install.
peek_agent_notes is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the peek_agent_notes 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 peek_agent_notes. 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.
peek_agent_notes is provided by the Claude Slack MCP server (theo-nash/claude-slack). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Claude Slack, 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.
19 Claude Slack tools catalogued and risk-classified — across an index of 43,000+ MCP servers.