Ai safety feature #23
Merged
Conversation
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
Co-authored-by: Shuyib <12908522+Shuyib@users.noreply.github.com>
Co-authored-by: Shuyib <12908522+Shuyib@users.noreply.github.com>
Co-authored-by: Shuyib <12908522+Shuyib@users.noreply.github.com>
…ecks and logging in Inspect AI layer
remove sudo and edit typos
format code with black
Added Gemma model recommendation and clarified VRAM requirements.
…nd project documentation
…1-f50d49393314 Add Inspect AI Safety Layer for Prompt Injection and Jailbreaking Detection
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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 pull request introduces significant improvements to the project's AI safety and model management workflows, focusing on integrating a robust Inspect AI safety layer, updating supported models, and enhancing Docker container startup reliability. The most important changes are grouped below by theme.
AI Safety Layer Integration
INSPECT_SAFETY_GUIDE.md) and integrated safety checks intoapp.py, evaluating user input for prompt injection, jailbreaking, and other attacks before LLM processing. Safety results are logged, and configuration options for strict/normal modes are provided. [1] [2] [3]README.mdto document the Inspect AI safety layer, its features, usage, integration points, and references, making security practices transparent and easy to follow.Model and Docker Management
qwen2.5:0.5btoqwen3:0.6b. Added a robust entrypoint script to ensure the model is pulled on container startup, improving reliability and startup diagnostics. [1] [2]docker-compose.ymlanddocker-compose-all.ymlto avoid conflicts and clarify external/internal port usage. [1] [2]Documentation and Project Structure
README.mdto include new recommended models (e.g., Gemma 27B), added references to the safety guide and implementation summary, and clarified file roles for easier onboarding and maintenance. [1] [2]Minor Fixes
Makefilefor improved clarity in linting instructions.These changes collectively strengthen the project's security posture, improve model support, and make deployment and usage more robust and transparent.