Store a fact, preference, or learning. Rules: - summary: one-line description (max 500 chars) — this is the index entry - content: full detail — be concise but complete - Always call memory_list first to check for existing similar memories - If updating a fact, memory_delete the old one then memo...
AI agents use memory_add to create or update resources in Shellgate — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Shellgate environment.
This tool creates and modifies data in a memory/storage system. It is Write category because it persistently stores information that can be created or updated. Severity is low because the worst case is storing incorrect facts or preferences, which does not affect external systems, finances, or result in irreversible data loss.
From the tool's definition Tool description explicitly states 'Store a fact, preference, or learning' and mentions 'memory_add the new' (creating/modifying entries). The operations are reversible via 'memory_delete' and do not cause financial, destructive, or code execution effects.
Documented attack patterns abuse exactly the kind of access memory_add gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Shellgate, and nothing reaches the server without passing your rules. This is the rule we recommend for memory_add:
{
"version": "1",
"default": "deny",
"tools": {
"memory_add": {
"limits": [
{
"counter": "memory_add_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}memory_add 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.
Store a fact, preference, or learning. Rules: - summary: one-line description (max 500 chars) — this is the index entry - content: full detail — be concise but complete - Always call memory_list first to check for existing similar memories - If updating a fact, memory_delete the old one then memory_add the new - One fact per memory — don. It is categorised as a Write tool in the Shellgate MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Shellgate MCP server in PolicyLayer and add a rule for memory_add: 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 Shellgate. Nothing to install.
memory_add 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 memory_add 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 memory_add. 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.
memory_add is provided by the Shellgate MCP server (matthiastjong/shellgate). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Shellgate, 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.
31 Shellgate tools catalogued and risk-classified — across an index of 43,000+ MCP servers.