Skip to content

feat(detection): add deterministic hidden-Unicode-obfuscation detection to triage #48

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
pengyuzhang merged 7 commits into from
Aug 26, 2026
+709 −18
Merged

feat(detection): add deterministic hidden-Unicode-obfuscation detection to triage #48

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
df8ed4a
feat(detection): add deterministic hidden-Unicode-obfuscation detecti…
Rahul-s-007 Aug 18, 2026
bf1b726
fix(detection): stop splicing decoded Unicode-obfuscation payload int…
Rahul-s-007 Aug 23, 2026
5ff1f51
test(detection): flag the pipeline coverage gap on the fixture test
Rahul-s-007 Aug 23, 2026
dfda5f6
fix(detection): also match escaped-literal Unicode-obfuscation payloads
Rahul-s-007 Aug 24, 2026
fc730ba
fix(detection): exclude U+E007F from the escaped-form Tag-Block regex
Rahul-s-007 Aug 24, 2026
354c4b4
fix(detection): recover the decoded Unicode-obfuscation payload into …
Rahul-s-007 Aug 26, 2026
283a373
Merge branch 'main' into feat/unicode-obfuscation-detector
pengyuzhang Aug 26, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 7 additions & 0 deletions Detection/context_providers/data/threat_repository.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -58,6 +58,13 @@ threat_framework:
- 'Malicious: Monitor MCP tool usage for data retrieval and analysis activities
that mimic legitimate processes but involve unauthorized access to sensitive
information.'
- 'Malicious: Monitor conversation and tool-output text for invisible Unicode
Tag-Block characters (U+E0020-U+E007E) that decode to hidden ASCII instructions,
an "ASCII smuggling" technique used to embed commands invisible to human
reviewers while remaining fully readable to the LLM.'
- 'Malicious: Monitor conversation and tool-output text for bidirectional-control
characters (U+202A-U+202E, U+2066-U+2069) used to visually hide or reorder
injected instructions from human reviewers while keeping the text machine-readable.'
- id: ADR.T0003
name: Agentic Control-Flow Hijacking
description: An adversary manipulates an agent to chain together individually
Expand Down
Loading