fix(compressor): soften summary prompt for content filters (salvage #19456)#21302
Merged
Conversation
Contributor
🔎 Lint report:
|
alt-glitch
added
type/bug
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
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes #19456 via salvage. Fixes #19362.
Summary
Context-compressor summary prompt used 'DIFFERENT assistant' / 'Do NOT respond' / 'injected as reference material' framing (inspired by OpenCode / Codex). Azure and OpenAI-compatible content filters flag that phrasing as prompt-injection-shaped and 400. Reword to plain 'structured checkpoint summary' / 'treat as source material' — same semantic goal, no injection-ish vocabulary. Structured checkpoint schema and safety constraints unchanged.
Validation
scripts/run_tests.sh tests/agent/test_context_compressor.py→ 69 passed (includes new regression inspecting the live prompt).Original author: @LeonSGP43.