Mercury

Overview

Mercury appears in the newsletter context as both a fintech company and a practical example of how modern product teams are applying AI to internal knowledge management and workflow automation. The strongest signal is from Mercury’s internal corpus: strategy docs, specs, queries, Slack threads, and transcripts were ingested into a local, QMD-indexed Claude Code knowledge base built from nearly 5 million words of company knowledge. This positions Mercury as a useful case study for AI Product Managers thinking about retrieval, enterprise context, and “second brain” systems for product work.

Mercury also shows up as an example of AI-ready product infrastructure. In the mentions, the company is connected to API-first thinking, MCP-based integrations, and external agent workflows such as OpenClaw. For AI PMs, Mercury matters less as a generic company profile and more as a concrete illustration of how product, data, and knowledge assets can be made accessible to AI systems through search, APIs, and secure connectors.

Key Developments

• 2026-02-08: Tal Raviv reportedly gave Opus 4.5 read-only access to a Mercury bank account through Mercury’s MCP connector, described as an official Anthropic app with quick OAuth, to diagnose a tax shortfall.
• 2026-04-08: Peter Yang referenced wiring Mercury, Google Workspace, and other APIs into his OpenClaw AI agent to automate the first 80% of docs, slides, and analytics workflows.
• 2026-04-23: Ryan Wiggs explained that Mercury prioritizes building robust APIs before MCPs, and described ingesting 5 million words from five years of PM work into Claude Code via QMD search to create a productivity-enhancing “second brain.”
• 2026-04-28: Ryan Wiggins was noted as having built a local QMD-indexed Claude Code knowledge base from nearly 5 million words of Mercury strategy docs, specs, queries, Slack threads, and transcripts.

Relevance to AI PMs

1. Blueprint for internal AI knowledge systems: Mercury is a strong example of how to turn fragmented institutional knowledge into a searchable AI layer. AI PMs can use this pattern to scope internal copilots for product strategy, specs, research, and decision history.

2. API-first before agent-first: The mention that Mercury builds robust APIs before MCPs is a practical lesson. AI PMs should ensure core product capabilities are exposed through stable APIs before layering on agent protocols, connectors, or chat-based interfaces.

3. Secure access and workflow automation: Mercury’s appearance in MCP and OAuth-based access flows highlights the operational side of AI products. PMs can study this as a model for permissioned read-only access, financial workflows, and safe enterprise integrations.

Related

• google-workspace: Mentioned alongside Mercury as part of an API set connected to OpenClaw for automating document, slide, and analytics work.
• openclaw: Peter Yang’s AI agent framework, used with Mercury and other APIs to automate a large share of knowledge work.
• peter-yang: Referenced Mercury in the context of agent-driven workflow automation and API integrations.
• tal-raviv: Connected Mercury to MCP-based account access and an AI-assisted financial diagnosis use case.
• opus-45: The model given read-only Mercury account access via MCP in the cited example.
• mcp: Mercury is referenced as having an MCP connector, making it relevant to AI tool interoperability.
• anthropic: Mercury’s MCP connector was described as an official Anthropic app in the newsletter mention.
• claude-code: The environment used to build a Mercury knowledge base from internal company materials.
• qmd: The indexing and search layer used to retrieve Mercury’s large internal corpus inside Claude Code.
• ryan-wiggs / ryan-wiggins: The person credited with describing and building the Mercury knowledge-base workflow.