{"slug": "three-traps-when-an-ai-assistant-reads-a-cpi-report", "title": "Three traps when an AI assistant reads a CPI report", "summary": "A developer connected Claude to a financial-data MCP server to analyze the October 2026 CPI report and identified three pitfalls when AI assistants interpret the data. The assistant correctly retrieves index levels for series like CPIAUCSL and CPILFESL, but can mishandle missing year-ago values left by the 2025 government shutdown, confuse seasonally adjusted with unadjusted indexes, and mislabel which month a release covers. The developer advises prompting the model to state which observation it used and to compare its year-over-year figure against the BLS release.", "body_md": "The October 2026 CPI report comes out today at 8:30 a.m. Eastern. I ran the run-up through Claude with a financial-data MCP server connected, and the tool calls are the easy part: `GetEconomicIndicator` returns the index levels for any CPI series, and the model does the division. The traps are in what the numbers mean. Here are three, with figures from a run on November 9 and 10, 2026. The full six-prompt walkthrough is in [How to Read a CPI Report With ChatGPT or Claude](https://equibles.com/blog/how-to-read-a-cpi-report-with-chatgpt-or-claude).\n\nFor September 2026, the latest month before today's release, headline CPI (`CPIAUCSL`) rose 0.6% on the month and 3.7% on the year. Core CPI (` CPILFESL`, all items less food and energy) rose 0.3% and 2.6%.\n\nAsk the assistant why, and it has to pull two more series. Energy (`CPIENGSL`) rose 5.2% in September and 20.4% over the year; food (` CPIUFDSL`) was flat on the month and up 2.5% on the year. That is most of the gap. A headline driven by energy can reverse as fast as it rose, which is why the prompt should always ask for core next to headline, and for the components when the two diverge.\n\nThese series have no October 2025 value. The Bureau of Labor Statistics did not publish one after the 2025 government shutdown. So when today's number lands, \"October 2026 versus October 2025\" has nothing to divide by.\n\nAn assistant that doesn't notice will do one of three things: skip to the nearest month, interpolate quietly, or report a change it can't have computed. Tell it to say which observation it used, and compare its year-over-year figure with the one in the BLS release.\n\nMonthly changes come from the seasonally adjusted index, while the official 12-month change uses the unadjusted one (`CPIAUCNS`). For September both give 3.7%, but they can differ by a tenth of a point, so ask which one the assistant used.\n\nThe label is the other trap. The report published on November 10 is for October; September's came out in October. \"The latest CPI\" before 8:30 a.m. and after it are different months, and a model answering from memory of the conversation can mix them up.\n\nTwo numbers put the release in context. The 10-year breakeven inflation rate was 2.22% on November 9, far below the 3.7% headline: bond traders treat this year's energy rise as temporary. And the VIX went from 16.61 on November 4 to 28.18 on November 5, then back to 19.92 on November 9, so the print lands on a market that was on edge a few days ago.\n\nThe October CPI is out. Compare headline and core with September, month over month and year over year, show the index values behind each change, say which month you used for the year-ago comparison, and split out energy and food.\n\nSetup for [Claude](https://equibles.com/docs/mcp/claude-web) or [ChatGPT](https://equibles.com/docs/mcp/chatgpt) takes a few minutes; the release schedule is on the [economic calendar](https://equibles.com/economicdata/calendar).\n\nSource: [Equibles](https://equibles.com/economicdata/cpiaucsl)", "url": "https://wpnews.pro/news/three-traps-when-an-ai-assistant-reads-a-cpi-report", "canonical_source": "https://dev.to/daniel_oliveira_4bd55d35d/three-traps-when-an-ai-assistant-reads-a-cpi-report-5i6", "published_at": "2026-11-10 06:23:57+00:00", "updated_at": "2026-11-10 06:25:17.474990+00:00", "lang": "en", "topics": ["ai-agents", "ai-tools", "large-language-models", "agent-protocols"], "entities": ["Claude", "ChatGPT", "Bureau of Labor Statistics", "Equibles", "Anthropic", "OpenAI"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/three-traps-when-an-ai-assistant-reads-a-cpi-report", "markdown": "https://wpnews.pro/news/three-traps-when-an-ai-assistant-reads-a-cpi-report.md", "text": "https://wpnews.pro/news/three-traps-when-an-ai-assistant-reads-a-cpi-report.txt", "jsonld": "https://wpnews.pro/news/three-traps-when-an-ai-assistant-reads-a-cpi-report.jsonld"}}