> For the complete documentation index, see [llms.txt](https://themochi-app.gitbook.io/mochi-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://themochi-app.gitbook.io/mochi-docs/mochi-sops/spam-filter-prompt-guide.md).

# Spam Filter Prompt Guide

### Why This Matters

Without a spam filter, your setters wade through PR pitches, editing service offers, follow-for-follow requests, and random DMs to find actual buyers. This destroys focus and inflates your lead numbers making your conversion data meaningless.

### Step-by-Step Setup

#### Step 1 — Collect Your Spam

For the first 3–5 days after connecting Mochi, screenshot every spam DM that comes in. Don't turn the filter on yet — collect the data first.\
Categories to look for:<br>

* Service pitches (editors, clippers, social media managers, web designers)
* Follow-for-follow or engagement pod requests
* Money requests / sob stories
* Generic compliments with no actual question
* Bot messages / emojis only
* People asking for free coaching or free content

#### Step 2 — Generate Your Prompt

Upload your screenshots to Claude or ChatGPT and use this prompt:\ <br>

```
Here are examples of spam DMs I receive on Instagram. Analyse the patterns and write me a spam filter prompt I can paste into a tool that uses AI to detect spam. 
Be specific and thorough. Include examples. The prompt should be broad enough to catch variations but specific enough not to filter legitimate leads who are genuinely interested in my offer.
```

#### Step 3 — Paste Into Mochi

Settings → Intelligence → Spam Filter → paste prompt → Save Changes<br>

#### Step 4 — Monitor for 7 Days

Check the Unqualified filter daily for the first week. If a legitimate lead got filtered:<br>

* Move them back to New Lead manually (one click)
* Add an exception to your spam filter prompt: *"Do NOT filter if the message contains \[specific phrase or intent]"*

### Example Spam Filter Prompt

Use this as a starting point and customise for your account:<br>

```
Messages matching any of the following patterns should be marked as spam. Apply these broadly — the examples are illustrative, not exhaustive.
SPAM SIGNALS:- Pitching a service (video editing, social media management,   copywriting, web design, clipping, thumbnails, ads, SEO, etc.)- Asking for a follow back, shoutout, or collaboration with no   genuine interest in the offer- Requesting free coaching, free content, or a free call   with no mention of budget or readiness to invest- Generic compliments with no question or real engagement   ("love your content", "you're so inspiring" with nothing else)- Requests for money, loans, or financial help- Messages that are emojis only or single words with no context- Obvious bot patterns — repetitive structure, no personalisation- People pitching their own product or service in the first message
DO NOT FILTER:- Anyone asking about the program, offer, pricing, or results- Anyone who references specific content (a post, reel, or story)- Anyone asking how to get started or book a call- Anyone who mentions a specific problem they're dealing with
```

### Common Mistakes

| <p>Mistake<br></p>                                           | <p>Fix<br></p>                                                            |
| ------------------------------------------------------------ | ------------------------------------------------------------------------- |
| <p>Turning on spam filter before building the prompt<br></p> | <p>Always build the prompt first — default filter is too generic<br></p>  |
| <p>Making the prompt too aggressive<br></p>                  | <p>Add explicit "do not filter" exceptions for real lead patterns<br></p> |
| <p>Never reviewing the Unqualified tab<br></p>               | <p>Check it weekly for the first month — catch false positives<br></p>    |
| <p>Using the same prompt forever<br></p>                     | <p>Review and refine quarterly as your audience evolves<br></p>           |


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://themochi-app.gitbook.io/mochi-docs/mochi-sops/spam-filter-prompt-guide.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.