metaharness_audit_trend
ADR-150 iter 15 — diff two oia-audit records (drift detection). Accepts EITHER memory keys (run metaharness_audit_list first to discover them) OR direct file paths (useful for diffing CI artifacts). Surfaces composite worst-severity delta + per-component status change + introduced/cleared finding...
This record as markdown: /tools/io-github-ruvnet-claude-flow/metaharness-audit-trend.md
What metaharness_audit_trend does on Claude Flow
AI agents call metaharness_audit_trend to retrieve information from Claude Flow without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Why metaharness_audit_trend is rated Low
The tool compares/diffs two existing audit records and surfaces differences — this is a read/query operation with no side effects. It reads from memory keys or file paths and reports findings. No data is written, executed, or deleted. Severity is low because misuse would only expose audit comparison results.
From the tool's definition diff two oia-audit records (drift detection)... Surfaces composite worst-severity delta + per-component status change + introduced/cleared findings + structural distance
Attacks that exploit this kind of access
The rule that runs metaharness_audit_trend safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Claude Flow, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For metaharness_audit_trend, this is the rule to start with:
metaharness_audit_trend is read-only, so it stays allowed. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Claude Flow, apply this rule, and every metaharness_audit_trend call is checked against it from then on.
Questions about metaharness_audit_trend
ADR-150 iter 15 — diff two oia-audit records (drift detection). Accepts EITHER memory keys (run metaharness_audit_list first to discover them) OR direct file paths (useful for diffing CI artifacts). Surfaces composite worst-severity delta + per-component status change + introduced/cleared findings + (iter 38) ADR-152 §3.1 structural distance when both records carry a fingerprint. Use when you have two specific audits to compare; pair with metaharness_audit_list for key discovery. Skipping this tool and eyeballing two JSONs is wrong because the structural-distance verdict (near-identical / minor-drift / moderate-drift / major-drift) is the operationally-useful summary. It is categorised as a Read tool in the Claude Flow MCP Server, which means it retrieves data without modifying state.
Register the Claude Flow MCP server in PolicyLayer and add a rule for metaharness_audit_trend: 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 Flow. Nothing to install.
metaharness_audit_trend 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_audit_trend 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_audit_trend. 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_audit_trend is provided by the Claude Flow MCP server (claude-flow). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Claude Flow, and thousands of servers like it.
This server
Across the catalogue