Home / Productivity · Agents / InBox Wall

InBox Wall
For safer AI email reading: locally strips hidden text, invisible Unicode and prompt injections.
What is InBox Wall
InBox Wall is a local-first, read-only email sanitizer written in Python. Before an AI reads your email, it removes hidden CSS text, invisible Unicode obfuscation and known prompt-injection patterns. The project is new and its repository does not declare a license.
How to use
- Install Python 3.10 or higher.
- Clone the repository: git clone https://github.com/Annora998/inboxwall.git, then cd inboxwall.
- Install the package and its dependencies: pip install -e .
- Check system health with inboxwall-doctor.
- Sync emails over read-only IMAP, with credentials stored in the OS keychain: inboxwall-sync Start the dashboard and local API with inboxwall-serve, or the MCP server for AI tools with inboxwall-mcp.
- Review quarantined emails in the dashboard's X-Ray view and choose Keep locked, Release after review or Report false positive.
Use cases
- Sanitizing incoming email HTML before an AI assistant reads it Stripping hidden text, invisible Unicode and known prompt-injection phrases from emails Serving only clean email text to AI tools through a local REST API or MCP server Quarantining and auditing suspicious emails locally with score explanations and an audit log
Price from official sources
Free
Check the official site for current pricing · Price not yet checked by our editors
Source: github.com · Info as of Nov 2, 2026
Reviews
No reviews yet.
Alternatives
-
sweet-search
Free
For agents searching codebases: local semantic search that saves 5–33% per agent run
- Added
- Nov 2, 2026
Same category: Productivity
-
nonce Not stated For securely granting AI access to your email: issues time-limited, least-privilege access per question
- Added
- Nov 2, 2026
Same category: Productivity
-
Skep Free For collaborating with coding agents across devices: share live sessions with a join code
- Added
- Nov 2, 2026
Same category: Productivity
-
Rufous
Free
For coding with small local models: an agent framework built for 2B–12B models
- Added
- Nov 2, 2026
Same category: Productivity
Social