mine_session_patterns
Analyze tool_call_log and learnings tables to extract recurring success/failure sequences across sessions. Finds common 2-3 tool sequences (bigrams/trigrams), cross-references with result_status to surface failure-prone patterns, and returns top patterns ranked by occurrence count and success rat...
This record as markdown: /tools/io-github-homenshum-nodebench/mine-session-patterns.md
What mine_session_patterns does on Nodebench
AI agents call mine_session_patterns to retrieve information from Nodebench without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Why mine_session_patterns is rated Low
This is primarily a Read operation—it retrieves and analyzes historical data from existing tables (tool_call_log, learnings, result_status) to identify patterns. However, the high severity reflects that the insights extracted could expose sensitive workflow patterns, failure modes, and success sequences that reveal how the system operates, which an adversary could exploit to craft better attacks or an unprivileged…
From the tool's definition Analyzes tool_call_log and learnings tables to extract recurring patterns and sequences, cross-referencing with result_status. Returns ranked patterns. The tool queries and inspects existing logs/tables without modifying them.
Attacks that exploit this kind of access
The rule that runs mine_session_patterns safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Nodebench, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For mine_session_patterns, this is the rule to start with:
mine_session_patterns 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 Nodebench, apply this rule, and every mine_session_patterns call is checked against it from then on.
Questions about mine_session_patterns
Analyze tool_call_log and learnings tables to extract recurring success/failure sequences across sessions. Finds common 2-3 tool sequences (bigrams/trigrams), cross-references with result_status to surface failure-prone patterns, and returns top patterns ranked by occurrence count and success rate. Use this to discover which tool workflows reliably succeed and which tend to fail. It is categorised as a Read tool in the Nodebench MCP Server, which means it retrieves data without modifying state.
Register the Nodebench MCP server in PolicyLayer and add a rule for mine_session_patterns: 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 Nodebench. Nothing to install.
mine_session_patterns 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 mine_session_patterns 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 mine_session_patterns. 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.
mine_session_patterns is provided by the Nodebench MCP server (nodebench-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Nodebench, and thousands of servers like it.
This server
Across the catalogue