Create documentation for Omega Walls integration#462
Open
anviren wants to merge 1 commit intodeepset-ai:mainfrom
Open
Create documentation for Omega Walls integration#462anviren wants to merge 1 commit intodeepset-ai:mainfrom
anviren wants to merge 1 commit intodeepset-ai:mainfrom
Conversation
This PR adds Omega Walls to the Haystack integrations index.
Omega Walls is an open-source runtime trust boundary for RAG and AI agent pipelines.
It includes a Haystack adapter, `OmegaHaystackGuard`, that can wrap a Haystack pipeline and guard tool calls before execution.
Use case:
Haystack pipelines often ingest external or semi-trusted content such as documents, PDFs, web pages, support tickets, knowledge base articles, or tool outputs. In LLM-powered workflows, that content is meant to be data, but it can still influence the model context and downstream tool behavior.
Omega Walls helps separate untrusted content from control flow by adding checks around:
- model/input flow
- tool calls before execution
- memory-write candidates
- cross-document or cross-step prompt-injection pressure
- secret-exfiltration pressure
- tool/action abuse
Install:
pip install "omega-walls[integrations]"
Haystack example:
from omega.integrations import OmegaHaystackGuard
guard = OmegaHaystackGuard(profile="quickstart")
pipeline = guard.wrap_pipeline(
pipeline,
component_name="omega_guard_component",
)
safe_tool = guard.wrap_tool("network_post", network_post_fn)
Links:
GitHub: https://github.com/synqratech/omega-walls
PyPI: https://pypi.org/project/omega-walls/
bilgeyucel
requested changes
May 11, 2026
Contributor
bilgeyucel
left a comment
bilgeyucel
left a comment
There was a problem hiding this comment.
Thanks for the PR @anviren! I left some comments. Feel free to add a logo for the integration as well
Comment on lines
+66
to
+71
| pipeline = guard.wrap_pipeline( | ||
| pipeline, | ||
| component_name="omega_guard_component", | ||
| ) | ||
|
|
||
| safe_tool = guard.wrap_tool("network_post", network_post_fn) |
Contributor
There was a problem hiding this comment.
Can I then use the pipeline/the tool as usual? What's the process after this step?
Comment on lines
+74
to
+105
| ### What it checks | ||
|
|
||
| Omega Walls is designed to check: | ||
|
|
||
| * model/input flow | ||
| * tool calls before execution | ||
| * memory-write candidates | ||
| * cross-document or cross-step prompt-injection pressure | ||
| * secret-exfiltration pressure | ||
| * tool/action abuse | ||
|
|
||
| When a risky input or tool call is detected, the guard can block the step with typed exceptions instead of letting the pipeline continue blindly. | ||
|
|
||
| ## When to use it | ||
|
|
||
| Omega Walls is useful for Haystack users building RAG or agentic workflows over real-world content, especially when the pipeline reads from: | ||
|
|
||
| * external documents | ||
| * PDFs | ||
| * web pages | ||
| * support tickets | ||
| * user-provided files | ||
| * internal knowledge bases | ||
| * tool outputs | ||
|
|
||
| Typical use cases include: | ||
|
|
||
| * internal document assistants | ||
| * support-ticket summarizers | ||
| * agentic RAG workflows | ||
| * pipelines with tool calls | ||
| * pipelines that read untrusted or user-provided documents |
Contributor
There was a problem hiding this comment.
Can you shorten this section in couple of sentences? Otherwise, it takes too much space
| --- | ||
| layout: integration | ||
| name: Omega Walls | ||
| description: Runtime trust boundary for Haystack RAG and agent pipelines that ingest untrusted content. |
Contributor
There was a problem hiding this comment.
Suggested change
| description: Runtime trust boundary for Haystack RAG and agent pipelines that ingest untrusted content. | |
| description: Runtime trust boundary for ingested content in Haystack applications |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR adds Omega Walls to the Haystack integrations index.
Omega Walls is an open-source runtime trust boundary for RAG and AI agent pipelines.
It includes a Haystack adapter,
OmegaHaystackGuard, that can wrap a Haystack pipeline and guard tool calls before execution.Use case:
Haystack pipelines often ingest external or semi-trusted content such as documents, PDFs, web pages, support tickets, knowledge base articles, or tool outputs. In LLM-powered workflows, that content is meant to be data, but it can still influence the model context and downstream tool behavior.
Omega Walls helps separate untrusted content from control flow by adding checks around:
Install:
pip install "omega-walls[integrations]"Haystack example:
Links:
GitHub: https://github.com/synqratech/omega-walls
PyPI: https://pypi.org/project/omega-walls/