Repository navigation
fix(runtime): bound complete SkillSpector requests before submission - #136
Merged
Merged
Conversation
…illspector-request-budgets # Conflicts: # docker/clawscan-runtime/patches/skillspector-responses.patch
|
🦞👀 Pull request received. I will update this pull request when review starts. |
This was referenced Nov 2, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Large bundled files can exceed SkillSpector’s model input budget and stop scanning before independent judgment. Carry the request-budget repair from NVIDIA/SkillSpector#980 onto our pinned integration, including packet schemas and metadata prompts.
Count the complete prompt, numbered code, relevant findings and selected response schema. Use an already-loaded codec or a conservative UTF-8 byte bound without downloading a vocabulary. Packet placement reserves schema and framing overhead, so a nearly full packet splits instead of discarding all packing. Unfit requests retain their inputs and fail before a provider call. Finding ownership, security checks, retries, deadlines and honest partial reporting remain intact.
The regression tests exercise both sync and async calls through plain pytest, with plugin autoload disabled. Native production qualification now targets the exact repaired runtime and locked 77/26 dependency sets.
Validation:
The exact DebugBundle 1.18.1 planner preserves all six files and 77,584 bundle lines under the unchanged 787,500 input budget. This model-free proof does not establish a completed scan or package acceptance. Both original-version scans, production adoption, publication and anonymous installation proof remain outstanding.