chore: remove all Anthropic API integration (never used)#73
Merged
Conversation
terse never used an Anthropic API key, so the anthropic code paths were dead weight that implied a workflow we don't run. Remove them entirely: - fluency.anthropic_answerer + dropeval.anthropic_tool_answerer (and the two _to_anthropic_* adapters) — the "real consumer" eval arm. The fluency eval reaches every model over the OpenAI-compatible path (broker pool or a loopback gateway). - tokenize.count_anthropic + measure --anthropic — the count_tokens "ground truth" check. cl100k stays the estimate; cross-tokenizer (cl100k vs o200k) invariance is the keyless robustness check, unchanged. - the --anthropic / --anthropic-model CLI flags, the Anthropic report section, the `anthropic` optional extra (pruned from uv.lock with its transitive deps), and the doc/env mentions. No behavior change to any path terse actually uses. 452 tests pass; ruff clean.
Accidentally staged by git add -A in the previous commit; it is a local symlink to the codegraph db, not repo content. Remove from tracking and gitignore it.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Removes all Anthropic API integration from terse — it was never used, so the code
paths were dead weight that implied a workflow we don't run.
Removed
fluency.anthropic_answerer+dropeval.anthropic_tool_answerer(and the_to_anthropic_tool/_to_anthropic_messagesadapters) — the "real consumer" evalarm. The fluency/drop evals reach every model over the OpenAI-compatible path
(broker pool or a loopback gateway); there's no other model backend.
tokenize.count_anthropic+measure --anthropic— thecount_tokens"ground truth" check. cl100k stays the estimate; cross-tokenizer (cl100k vs o200k)
invariance is the keyless robustness check, unchanged.
--anthropic/--anthropic-modelCLI flags, the Anthropic section of themeasure report, the
anthropicoptional extra (pruned fromuv.lockwith itstransitive deps: httpx/anyio/distro/…), and the doc +
ANTHROPIC_API_KEYenv mentions..codegraphsymlink and gitignore it.Verification
ruffclean,uv lock --checkconsistent.terse fluency --help/terse measure --helpcarry no--anthropicflag; allmodules import.
Anthropic key that was never present.
Independent of #71/#72 (fluency nested-record coverage); this is pure removal.