Repository navigation
fix(library)!: reject inputs exceeding detector coverage in jailbreak detection - #2560
Merged
Merged
Conversation
github-actions
Bot
added
status: needs triage
Signed-off-by: Pouyanpi <13303554+Pouyanpi@users.noreply.github.com>
Signed-off-by: Pouyanpi <13303554+Pouyanpi@users.noreply.github.com>
Pouyanpi
force-pushed
the
pouyanpi/jailbreak-truncation-bypass
branch
from
November 4, 2026 10:55
98f575e to
664390c
Compare
Contributor
|
Staged Fern docs preview: https://nvidia-preview-pr-2560.docs.buildwithfern.com/nemo/guardrails |
Signed-off-by: Pouyanpi <13303554+Pouyanpi@users.noreply.github.com>
Pouyanpi
force-pushed
the
pouyanpi/jailbreak-truncation-bypass
branch
from
November 4, 2026 11:25
664390c to
c07f1d6
Compare
tgasser-nv
approved these changes
Nov 6, 2026
tgasser-nv
left a comment
tgasser-nv
left a comment
Collaborator
There was a problem hiding this comment.
Looks good, just a few nits to clean up before merging. High level item:
- Should the PR description be
fix(library)!: ....since prior to the PR inputs > 2,048 tokens were classified using the first 2,048 tokens and now they'll be rejected before the model runs?
Signed-off-by: Pouyanpi <13303554+Pouyanpi@users.noreply.github.com>
Signed-off-by: Pouyanpi <13303554+Pouyanpi@users.noreply.github.com>
BREAKING CHANGE: The jailbreak model rail rejects messages over 2048 Snowflake tokens, including special tokens, instead of classifying only their truncated prefix. Signed-off-by: Pouyanpi <13303554+Pouyanpi@users.noreply.github.com>
Pouyanpi
changed the title
Pouyanpi
removed
the
status: needs triage
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
Block inputs over 2,048 tokens, including special tokens, before cache lookup or local/remote classification. Use a bundled, pinned Snowflake tokenizer for offline validation, and block on validation failures or detector HTTP 413 responses.
This protects calls through Guardrails; direct calls to the hosted NIM still depend on its own input handling.
Related Issue(s)
Verification