basitalisandhu

basitalisandhu/spotlighting

Spotlighting for prompt injection defence in Python: marks untrusted text before it reaches a model with delimiting, datamarking and encoding as described in the Microsoft Research paper, plus the system-prompt instruction for each mode. Pipeline for tool results, helpers for web pages and email, CLI. Standard library only, Python 3.11+.

6 good first / help-wanted issues · Python · last activity Oct 4, 2026

0 stars 0 forks 0 watchers Python MIT License
agent-security agentic-ai ai-agents cli datamarking guardrails hacktoberfest indirect-prompt-injection llm-security prompt-injection prompt-injection-defense python spotlighting untrusted-input
View on GitHub Website
6 Open Issues Need Help Last updated: Oct 4, 2026

Open Issues Need Help

View All on GitHub
Add a task option to the instruction text 20 days ago
help wanted good first issue hacktoberfest
basitalisandhu/spotlighting
0

Spotlighting for prompt injection defence in Python: marks untrusted text before it reaches a model with delimiting, datamarking and encoding as described in the Microsoft Research paper, plus the system-prompt instruction for each mode. Pipeline for tool results, helpers for web pages and email, CLI. Standard library only, Python 3.11+.

Python
#agent-security#agentic-ai#ai-agents#cli#datamarking#guardrails#hacktoberfest#indirect-prompt-injection#llm-security#prompt-injection#prompt-injection-defense#python#spotlighting#untrusted-input
Add include_title option to html_to_text 21 days ago
help wanted good first issue hacktoberfest
basitalisandhu/spotlighting
0

Spotlighting for prompt injection defence in Python: marks untrusted text before it reaches a model with delimiting, datamarking and encoding as described in the Microsoft Research paper, plus the system-prompt instruction for each mode. Pipeline for tool results, helpers for web pages and email, CLI. Standard library only, Python 3.11+.

Python
#agent-security#agentic-ai#ai-agents#cli#datamarking#guardrails#hacktoberfest#indirect-prompt-injection#llm-security#prompt-injection#prompt-injection-defense#python#spotlighting#untrusted-input
Add a for_document helper for plain files 21 days ago
help wanted good first issue hacktoberfest
basitalisandhu/spotlighting
0

Spotlighting for prompt injection defence in Python: marks untrusted text before it reaches a model with delimiting, datamarking and encoding as described in the Microsoft Research paper, plus the system-prompt instruction for each mode. Pipeline for tool results, helpers for web pages and email, CLI. Standard library only, Python 3.11+.

Python
#agent-security#agentic-ai#ai-agents#cli#datamarking#guardrails#hacktoberfest#indirect-prompt-injection#llm-security#prompt-injection#prompt-injection-defense#python#spotlighting#untrusted-input
Support a ** wildcard in trusted_paths 21 days ago
help wanted good first issue hacktoberfest
basitalisandhu/spotlighting
0

Spotlighting for prompt injection defence in Python: marks untrusted text before it reaches a model with delimiting, datamarking and encoding as described in the Microsoft Research paper, plus the system-prompt instruction for each mode. Pipeline for tool results, helpers for web pages and email, CLI. Standard library only, Python 3.11+.

Python
#agent-security#agentic-ai#ai-agents#cli#datamarking#guardrails#hacktoberfest#indirect-prompt-injection#llm-security#prompt-injection#prompt-injection-defense#python#spotlighting#untrusted-input
Add line wrapping to base64 encode 21 days ago
help wanted good first issue hacktoberfest
basitalisandhu/spotlighting
0

Spotlighting for prompt injection defence in Python: marks untrusted text before it reaches a model with delimiting, datamarking and encoding as described in the Microsoft Research paper, plus the system-prompt instruction for each mode. Pipeline for tool results, helpers for web pages and email, CLI. Standard library only, Python 3.11+.

Python
#agent-security#agentic-ai#ai-agents#cli#datamarking#guardrails#hacktoberfest#indirect-prompt-injection#llm-security#prompt-injection#prompt-injection-defense#python#spotlighting#untrusted-input
Add --with-instruction to spotlighting mark 21 days ago
help wanted good first issue hacktoberfest
basitalisandhu/spotlighting
0

Spotlighting for prompt injection defence in Python: marks untrusted text before it reaches a model with delimiting, datamarking and encoding as described in the Microsoft Research paper, plus the system-prompt instruction for each mode. Pipeline for tool results, helpers for web pages and email, CLI. Standard library only, Python 3.11+.

Python
#agent-security#agentic-ai#ai-agents#cli#datamarking#guardrails#hacktoberfest#indirect-prompt-injection#llm-security#prompt-injection#prompt-injection-defense#python#spotlighting#untrusted-input