Switch the observation scope to a different git branch. Observations are scoped per-branch.
AI agents use switch_context to create or update resources in Codewatch Memory — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Codewatch Memory environment.
This tool changes the active context/scope of observations from one git branch to another. It modifies the server's state (which branch is currently active for scoping observations), which is a reversible write/configuration change. It does not delete data, execute code, or involve financial operations.
From the tool's definition Switch the observation scope to a different git branch. Observations are scoped per-branch.
Attacks that exploit this kind of access
Switch the observation scope to a different git branch. Observations are scoped per-branch. It is categorised as a Write tool in the Codewatch Memory MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Codewatch Memory MCP server in PolicyLayer and add a rule for switch_context: 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 Codewatch Memory. Nothing to install.
switch_context 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 switch_context 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 switch_context. 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.
switch_context is provided by the Codewatch Memory MCP server (klausandrade/codewatch). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →