Break Trace

Traces a reconciliation break back to the source GL entry and subledger transaction, then writes a one-sentence diagnosis explaining what differs and why. It pulls posting metadata from both sides via MCP servers, diffs attributes like FX rates, posting dates, and account mappings, and outputs a structured root cause with an owner and recommended action. Use this after gl-recon has flagged the break but before you write any adjustments. The real value is speed: instead of manually pulling trade IDs and journal entries from two systems, you get "Subledger used WM/R 4pm rate; GL used Bloomberg close" in one shot. It won't fix the break, but it will tell you whether to monitor, escalate, or suppress.

skills.sh installs: 1.3K
GitHub stars: 41.2K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add anthropics/financial-services --skill break-trace --agent claude-code

Installs into .claude/skills of the current project.

Files

Root-cause a break

Given a single break row (key, GL values, subledger values, bucket, likely cause), trace it to source and produce a root-cause statement.

Trace path

Pull the GL side — via the internal-gl MCP, fetch the journal entry or posting that produced this GL line: entry id, posting date, source system, batch id, preparer.
Pull the subledger side — via the subledger MCP, fetch the matching transaction: trade id, trade/settle dates, counterparty, source feed, FX rate used.
Diff the attributes — line up posting date, FX rate/date, account mapping, quantity sign, amount sign. The differing attribute is usually the cause.