chore: align Sinapsis with Claude Opus 4.8 (v4.6.0)#19
Merged
Conversation
Opus 4.8 keeps the same 1M context as 4.7 but holds long context with fewer compactions and better recovery, lowers the minimum cacheable prompt to 1,024 tokens, and accepts mid-conversation system messages -- the exact shape of the per-turn systemMessage injection Sinapsis already does. - Re-tune deterministic hot-path caps: MAX_INSTINCTS_INJECTED 6 -> 8, TOKEN_BUDGET 4000 -> 6000, learner observation window 5000 -> 8000 lines. - Retarget docs/rfc-v5-adaptive-thinking.md to claude-opus-4-8 with adaptive thinking + effort (budget_tokens returns 400 on 4.7+); fix the cache note (an ~800-token block is below the 1,024-token cache minimum on 4.8). - Bump to v4.6.0 (README header + badge), add CHANGELOG and README entries, note v4.6 caps in FEATURES. - Add tests/test-v46-opus48.sh (8 tests). All suites green: v46 8/8, v45 11/11, security 11/11, gstack 18/18, v433 14/14, dream 25/25. Hot path stays model-free; effort default is high in Claude Code, no change needed. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Opus 4.8 alignment (v4.6.0)
Opus 4.8 keeps the same 1M context as 4.7 but holds long context with fewer compactions and better recovery, lowers the minimum cacheable prompt to 1,024 tokens, and accepts mid-conversation
systemmessages — the exact shape of Sinapsis's per-turnsystemMessageinjection.Changed
MAX_INSTINCTS_INJECTED6 → 8,TOKEN_BUDGET4000 → 6000, learner window 5000 → 8000 lines. Justified by 4.8's better long-context handling, not a bigger window (1M unchanged).docs/rfc-v5-adaptive-thinking.mdretargeted toclaude-opus-4-8with adaptive thinking +effort; corrected the cache note (an ~800-token block is below 4.8's 1,024-token cache minimum and would not cache).Not changed
effortdefaults tohighin Claude Code; Sinapsis never calls the model itself.Tests
tests/test-v46-opus48.sh(8 tests: caps, model-ID alignment, version bump).🤖 Generated with Claude Code