DisclaimerUnofficial, and not affiliated with Anthropic. Nearly all of this is read straight out of what ships: npm bundles, captured prompts, published docs. Anthropic's own notes go in verbatim, marked as theirs. The rest is my reading, and every entry carries the strings behind it. If one looks wrong, vote it down and say why.
Improved Hook Evaluator: Transcript Truncation and Model Awareness
What's wrong with this entry?
Anonymous. No account, no email.
Hook evaluators now truncate conversation transcripts to fit within the evaluator's context window and pass the correct model identifier. When truncation occurs, a marker is inserted: "[Earlier conversation truncated to fit the hook evaluator's context window — N earlier messages omitted...]"
Evidence
Function zQY with truncation logic (search for "Earlier conversation truncated to fit the hook evaluator's context window")
Strings lifted out of the shipped bundle, so the claim above can be checked against them.