Bounce Monitoring Workflow logo

Bounce Monitoring Workflow

Community
TomGranot
bounce-monitoring-workflow

Build a workflow to protect sender reputation through automated bounce monitoring. Auto-suppresses contacts above a configurable bounce threshold, alerts on hard bounces, and flags high-bounce contacts for weekly manual review.

Overview

PublisherTomGranot
Repositoryhubspot-admin-skills
Skill namebounce-monitoring-workflow
Stars
79
Forks
18
Bundled files
3
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 3 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by TomGranot on GitHub. Read the source before you install it.

Installation

Install the Bounce Monitoring Workflow AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/TomGranot/hubspot-admin-skills/tree/main/skills/bounce-monitoring-workflow
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/TomGranot/hubspot-admin-skills.git /tmp/hubspot-admin-skills
mkdir -p .claude/skills
cp -r /tmp/hubspot-admin-skills/skills/bounce-monitoring-workflow .claude/skills/bounce-monitoring-workflow
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Bounce Monitoring Workflow in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Bounce Monitoring Workflow on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Bounce Monitoring Workflow is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Bounce Monitoring Workflow

Protect your email sender reputation with automated bounce detection and suppression. This workflow catches bounces as they happen rather than waiting for periodic cleanup.

How It Works

ConditionAction
Hard bounce detectedAlert admin immediately, suppress contact
Suppression threshold reached (commonly 2-3 bounces)Auto-suppress from marketing emails
Review threshold reached (commonly 3-5 bounces)Flag for weekly manual review (delete vs. recover)

Prerequisites

  • HubSpot Marketing Professional or Enterprise plan
  • A HubSpot service key (HUBSPOT_ACCESS_TOKEN in .env; an existing legacy private app token also works) with the automation scope (for the API path)
  • Python 3.10+ with uv
  • A custom contact property for review flagging (default: email_health_flag — the execute script creates it if missing)
  • Admin email or Slack channel for hard bounce alerts

Scripts

Script paths are relative to this skill's directory. Run them from your project directory so .env, data/, and reports/ resolve there: prefix each path with wherever this skill is installed (skills/bounce-monitoring-workflow/ in a clone of the repo).

StageScriptRun with
Beforescripts/before.pyuv run scripts/before.py
Executescripts/execute.pyuv run scripts/execute.py
Afterscripts/after.pyuv run scripts/after.py

before.py inventories existing workflows and checks name collisions. execute.py creates the email_health_flag property (if missing) and three bounce workflows via POST /automation/v4/flows — always disabled, for review before enabling. after.py verifies and reports enabled state.

Building the Workflow: Two Options

Option A: Create via the v4 Automation API (primary)

API status: HubSpot documents the Workflows API as beta and subject to change. Its current reference is the date-based beta path /automation/v4/2027-03-beta/flows; this skill's scripts still call the unversioned /automation/v4/flows path, which v4's end of support on March 30, 2027 also covers. Creating workflows disabled and reviewing them in the UI is what makes a beta API acceptable here. If a script's create call starts failing, the manual UI build below is the fallback.

Instead of one workflow with three nested branches, the script decomposes the design into three linear workflows — the same coverage, cleanly expressible via the API and independently tunable:

  1. Hard Bounce Suppression: enrollment = hs_email_hard_bounce_reason_enum is known. Actions: set email_health_flag, set marketing contact status to non-marketing.
  2. Suppress at N+ Bounces: enrollment = hs_email_bounce >= suppression threshold (re-enrollment on). Action: set marketing contact status to non-marketing.
  3. Flag for Review at M+ Bounces: enrollment = hs_email_bounce >= review threshold (re-enrollment on). Action: set email_health_flag (feeds /review-bounced-contacts).

Configure SUPPRESSION_THRESHOLD and REVIEW_THRESHOLD in execute.py, then:

bash
uv run scripts/before.py
uv run scripts/execute.py

Safety model: all three workflows are created with isEnabled: false. During UI review: add the internal-notification actions (hard bounce alert, review alert — notification recipients are portal-specific), add "Set marketing contact status" manually if the script reported it had to omit it, then turn the workflows on.

Option B: Manual UI Build (single workflow with nested branches)

Follow the step-by-step instructions in Stage 3 below — the original single-workflow design with three branch levels. Use this on portals where the Automation API is unavailable.

Alternatives: HubSpot Breeze AI can scaffold a similar workflow from a prompt, but it creates event-based (OR) triggers, cannot configure re-enrollment, and may flatten nested branches — verify everything it builds. The Claude Chrome extension can drive the workflow builder UI directly. Both are secondary options.

Step-by-Step Build Instructions

Stage 1: Plan

  1. Choose your suppression threshold (commonly 2-3 bounces) and review threshold (commonly 3-5).
  2. Decide who receives hard-bounce and review alerts.

Stage 2: Before

  1. Run uv run scripts/before.py to inventory workflows and check name collisions.
  2. Identify your current bounce baseline — run a quick search for contacts where hs_email_bounce > 0 to understand the starting volume.
  3. If building manually, create your bounce review property (checkbox or dropdown) — the API path creates it automatically.

Stage 3: Execute

Option A: run uv run scripts/execute.py, then complete the UI review steps listed above.

Option B — manual UI build: build a single contact-based workflow with branching logic.

  1. Trigger: hs_email_bounce is known (fires when bounce count updates)

  2. Branch 1: Hard bounce check

    • Condition: hs_email_hard_bounce_reason_enum is known
    • YES:
      • Send internal notification: "Hard bounce: {email} — {hs_email_hard_bounce_reason_enum}"
      • Set hs_marketable_status to non-marketing (workflow action)
    • NO: Continue to Branch 2
  3. Branch 2: Bounce count >= your suppression threshold (commonly 2-3)

    • Condition: hs_email_bounce is greater than or equal to your suppression threshold
    • YES:
      • Set hs_marketable_status to non-marketing (workflow action)
      • Continue to Branch 3
    • NO: No action (below threshold — monitor only)
  4. Branch 3: Bounce count >= your review threshold (commonly 3-5)

    • Condition: hs_email_bounce is greater than or equal to your review threshold
    • YES:
      • Set your bounce review property = flagged
      • Send internal notification: "Contact {email} has [review threshold]+ bounces — review for deletion"
    • NO: No further action
  5. Settings:

    • Re-enrollment: ON (contact should re-enter if bounce count increases)
    • Goal: None
  6. Turn on the workflow.

Stage 4: After

  1. Run uv run scripts/after.py (API path) to confirm the workflows exist and report enabled state.
  2. Check workflow history after the first week of email sends.
  3. Review the flagged contacts list weekly — decide for each contact:
    • Delete if the email is clearly invalid (typo domain, defunct company)
    • Attempt recovery if the domain is valid (could be a temporary mailbox issue)
  4. Monitor overall bounce rate in HubSpot email health dashboard.

Rollback

  1. Turn off the workflow.
  2. Contacts already suppressed remain non-marketing. To reverse:
    • Filter contacts suppressed by this workflow (check your bounce review property or workflow history)
    • Manually set back to marketing contacts in the UI
  3. Clear your bounce review property values in bulk if needed.

Weekly Review Process

For contacts flagged at your review threshold:

  1. Export the flagged contacts list.
  2. For each contact, check:
    • Is the email domain still active? (Quick MX record check)
    • Is this a known customer or high-value contact?
    • Was the bounce recent or historical?
  3. Delete contacts with invalid domains or clearly fake emails.
  4. Keep suppressed contacts with valid domains but repeated soft bounces.
  5. Clear the bounce review property after review.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Bounce Monitoring Workflow AI skill do?

Build a workflow to protect sender reputation through automated bounce monitoring. Auto-suppresses contacts above a configurable bounce threshold, alerts on hard bounces, and flags high-bounce contacts for weekly manual review.

Why use Bounce Monitoring Workflow on TypingMind?

Because you install it once and use it with any model. Bounce Monitoring Workflow is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Bounce Monitoring Workflow in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/TomGranot/hubspot-admin-skills/tree/main/skills/bounce-monitoring-workflow. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Bounce Monitoring Workflow?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Bounce Monitoring Workflow?

As many as you like. As long as a model supports skills, you can use Bounce Monitoring Workflow with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Bounce Monitoring Workflow AI skill free?

Yes. It is published on GitHub by TomGranot under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇