back_office_ops · workflow
Mercari PM intern builds mercari-pm-agent automating end-to-end PM workflow via Claude Code Skills and MCP
Mercari PMs needed to gather information from Notion, Slack, Looker, and Figma before every product decision, and the time spent moving across tools and organizing which data mattered was taking PMs away from deeper thinking and stakeholder conversations.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · PM describes business problem
When a PM describes a business problem on the product in natural language, the workflow steps proceed automatically.
Tools used
Claude CodeNotion MCPSlack MCPSocratesFigma MCPBigQueryLooker · partner
Outcome
Separating SKILL.md from reference files produced a clear improvement in evaluation score, and the agent can now run the full PM workflow end-to-end within a single session.
What failed first
Consolidating all agent behavior definitions into a single SKILL.md file produced poor output accuracy due to the LLM 'Lost in the Middle' problem, where models fail to attend to relevant information in long contexts.
Results
Time savedreduces waiting time compared to sequential access
Volumeclear improvement in score
Grounding & classification
Source type: technical build writeup
24 fields verified against source quotes, 2 dropped as unverifiable.
agentic workflowcontent generationdata extractionmulti agent workflowchat transcriptknowledge basebuilder submittedhuman review describednamed customertools describedecommercesoftwareemployee productivitytime savedtechnical build writeupback office opsagentic task executiondocument to record