ADR-152 §3.1 — weighted similarity between two harness fingerprints (genome + score JSON). Returns overall ∈ [0,1] plus per-component breakdown (cosine over 9 numerics, categorical over 4 enums, jaccard over agent_topology). Pure-TS, zero
AI agents call metaharness_similarity to retrieve information from Ruflo without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This is a comparison/analysis function that retrieves and computes metrics over existing fingerprint data without modifying any state, executing external operations, or affecting other systems. It fits the Read category: retrieves and analyzes data with no side effects. Low severity due to purely informational nature.
From the tool's definition Tool computes and returns 'weighted similarity between two harness fingerprints' with output of numeric scores and breakdowns. The description emphasizes 'Pure-TS, zero' (likely 'zero side effects').
Attacks that exploit this kind of access
ADR-152 §3.1 — weighted similarity between two harness fingerprints (genome + score JSON). Returns overall ∈ [0,1] plus per-component breakdown (cosine over 9 numerics, categorical over 4 enums, jaccard over agent_topology). Pure-TS, zero. It is categorised as a Read tool in the Ruflo MCP Server, which means it retrieves data without modifying state.
Register the Ruflo MCP server in PolicyLayer and add a rule for metaharness_similarity: 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 Ruflo. Nothing to install.
metaharness_similarity 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 metaharness_similarity 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 metaharness_similarity. 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.
metaharness_similarity is provided by the Ruflo MCP server (ruvnet/ruflo). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
metaharness_similarity is one line of Ruflo's registry record.
The record carries the whole server: verified identity, auth posture, risk grade, every tool classified, recommended policy — re-checked continuously.
Teams ship this data inside their own products. See what a licence covers →