← All agents

AI & research · One purpose, multiple implementations

3.2: Fetch Last Email Correspondance

Adapt the published “3.2: Fetch Last Email Correspondance” workflow to other implementations while preserving its documented behavior.

The source contains 47 executable nodes and 47 connections. This draft is derived from the n8n export; it needs a human-reviewed purpose, input/output contract, and behavioral tests before a port can claim equivalence.

Open agent files on GitHub ↗

Shared workflow

This initial Mermaid diagram shows observed connections from the n8n source, including model and tool dependencies. Unresolved connections are listed under Evidence. It is a starting point for the shared design; it does not execute the agent or establish correctness.

Rendering the workflow diagram…

Read the Mermaid source
flowchart LR
  n0["Get Message Contents"]
  n1["Simplify Emails"]
  n2["Check For Upcoming Meetings"]
  n3["OpenAI Chat Model2"]
  n4["Extract Attendee Information"]
  n5["Execute Workflow Trigger"]
  n6["OpenAI Chat Model"]
  n7["Get Last Correspondence"]
  n8["OpenAI Chat Model1"]
  n9["OpenAI Chat Model3"]
  n10["WhatsApp Business Cloud"]
  n11["Schedule Trigger"]
  n12["Return LinkedIn Success"]
  n13["Return LinkedIn Error"]
  n14["Return Email Error"]
  n15["Return Email Success"]
  n16["Set Route Email"]
  n17["Set Route Linkedin"]
  n18["Router"]
  n19["Return LinkedIn Error1"]
  n20["Has Emails#63;"]
  n21["Return Email Error1"]
  n22["Sections To List"]
  n23["Set LinkedIn Cookie"]
  n24["Attendees to List"]
  n25["Merge Attendee with Summaries"]
  n26["Has Email Address#63;"]
  n27["Has LinkedIn URL#63;"]
  n28["Get Correspondance"]
  n29["Merge"]
  n30["Aggregate Attendees"]
  n31["Activities To Array"]
  n32["Extract Profile Metadata"]
  n33["Activities To List"]
  n34["APIFY Web Scraper"]
  n35["Get Activity Details"]
  n36["Get Sections"]
  n37["Get About Section"]
  n38["Get Activity Section"]
  n39["Extract Activities"]
  n40["Merge1"]
  n41["Is Scrape Successful#63;"]
  n42["Extract About"]
  n43["Get LinkedIn Profile #38; Activity"]
  n44["Correspondance Recap Agent"]
  n45["Attendee Research Agent"]
  n46["LinkedIn Summarizer Agent"]
  n29 -->|"main out 0 to in 0"| n25
  n40 -->|"main out 0 to in 0"| n46
  n18 -->|"main out 0 to in 0"| n26
  n18 -->|"main out 1 to in 0"| n27
  n20 -->|"main out 0 to in 0"| n0
  n20 -->|"main out 1 to in 0"| n14
  n36 -->|"main out 0 to in 0"| n37
  n36 -->|"main out 0 to in 0"| n38
  n42 -->|"main out 0 to in 0"| n40
  n16 -->|"main out 0 to in 0"| n28
  n1 -->|"main out 0 to in 0"| n44
  n11 -->|"main out 0 to in 0"| n2
  n22 -->|"main out 0 to in 0"| n36
  n34 -->|"main out 0 to in 0"| n41
  n24 -->|"main out 0 to in 0"| n16
  n24 -->|"main out 0 to in 0"| n17
  n37 -->|"main out 0 to in 0"| n42
  n27 -->|"main out 0 to in 0"| n23
  n27 -->|"main out 1 to in 0"| n19
  n6 -->|"ai_languageModel out 0 to in 0"| n44
  n33 -->|"main out 0 to in 0"| n35
  n39 -->|"main out 0 to in 0"| n33
  n28 -->|"main out 0 to in 0"| n29
  n26 -->|"main out 0 to in 0"| n7
  n26 -->|"main out 1 to in 0"| n21
  n8 -->|"ai_languageModel out 0 to in 0"| n46
  n3 -->|"ai_languageModel out 0 to in 0"| n4
  n9 -->|"ai_languageModel out 0 to in 0"| n45
  n17 -->|"main out 0 to in 0"| n43
  n31 -->|"main out 0 to in 1"| n40
  n30 -->|"main out 0 to in 0"| n45
  n23 -->|"main out 0 to in 0"| n34
  n35 -->|"main out 0 to in 0"| n31
  n38 -->|"main out 0 to in 0"| n39
  n0 -->|"main out 0 to in 0"| n1
  n41 -->|"main out 0 to in 0"| n32
  n41 -->|"main out 1 to in 0"| n13
  n45 -->|"main out 0 to in 0"| n10
  n7 -->|"main out 0 to in 0"| n20
  n5 -->|"main out 0 to in 0"| n18
  n32 -->|"main out 0 to in 0"| n22
  n46 -->|"main out 0 to in 0"| n12
  n44 -->|"main out 0 to in 0"| n15
  n2 -->|"main out 0 to in 0"| n4
  n4 -->|"main out 0 to in 0"| n24
  n25 -->|"main out 0 to in 0"| n30
  n43 -->|"main out 0 to in 1"| n29
Read the specification and unresolved requirements ↗

Implementations

Each implementation shares the agent’s purpose. Available source, planned ports, and verified results are labeled separately.

Plain Python

Planned · no code yet

Read the implementation plan ↗

Pydantic AI

Planned · no code yet

Read the implementation plan ↗

CrewAI Flow

Planned · no code yet

Read the implementation plan ↗

CrewAI Crew

Planned · no code yet

Read the implementation plan ↗

LangGraph

Planned · no code yet

Read the implementation plan ↗

Build or operate with an assistant

Each prompt includes separate build and operating instructions tailored to this agent. They are untested instructions, not working implementations. Check the chosen product’s actual capabilities before using it.

Evidence and remaining work

The bulk collection has publication checks, not live-service testing. Diagrams and prompts are documentation. A port needs reviewed schemas, fixtures, actual execution results, and explicit limits before claiming equivalent behavior.

  • Define input/output schemas and expected outputs.
  • Review source expressions, external actions, and failure behavior.
  • Determine whether model judgment is required or deterministic rules suffice.

Curated by Nik Bear Brown. Preserve each implementation’s original attribution and license. These source templates are separate from the audited Claude directory.