Record an interaction outcome to update trust scores. Valid outcomes: success, failure, timeout, partial. Positive outcomes raise trust; negative outcomes lower it. Args: peer_did: The DID of the peer agent involved. outcome: Interaction result — one of success, failure, timeout, partial. details...
AI agents use record_interaction to create or update resources in Agentos — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Agentos environment.
This tool creates/writes new records and modifies existing trust score state in a reversible manner. It does not delete data (Destructive), does not directly execute code or external operations (Execute), and does not involve financial transactions (Financial).
From the tool's definition Tool modifies state by recording interaction outcomes that 'update trust scores'. The description explicitly indicates this creates or updates records ('Record an interaction outcome').
Documented attack patterns abuse exactly the kind of access record_interaction gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Agentos, and nothing reaches the server without passing your rules. This is the rule we recommend for record_interaction:
{
"version": "1",
"default": "deny",
"tools": {
"record_interaction": {
"limits": [
{
"counter": "record_interaction_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}record_interaction 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.
Record an interaction outcome to update trust scores. Valid outcomes: success, failure, timeout, partial. Positive outcomes raise trust; negative outcomes lower it. Args: peer_did: The DID of the peer agent involved. outcome: Interaction result — one of success, failure, timeout, partial. details: Free-text description of the interaction. It is categorised as a Write tool in the Agentos MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Agentos MCP server in PolicyLayer and add a rule for record_interaction: 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 Agentos. Nothing to install.
record_interaction 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 record_interaction 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 record_interaction. 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.
record_interaction is provided by the Agentos MCP server (@microsoft/agentos-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Agentos, 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.
18 Agentos tools catalogued and risk-classified — across an index of 43,000+ MCP servers.