Align agents-guide differentiation paragraph with Kitaru replay messaging#5043
Open
AlexejPenner wants to merge 6 commits into
Open
Align agents-guide differentiation paragraph with Kitaru replay messaging#5043AlexejPenner wants to merge 6 commits into
AlexejPenner wants to merge 6 commits into
Conversation
…non v3 Name Braintrust/LangSmith/Langfuse, state the watched-vs-executed distinction, and keep the single-shot honest boundary. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Drops the "only Kitaru executed your agent" claim (the harness executes the agent; Kitaru is the runtime it ran on), stops implying Langfuse can test against a dataset, and removes "genuinely" from the boundary line. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Competitors are passive (they ingest run data), and the replay-semantics sentence now matches how replay actually works: model calls re-run for real; tool calls answer from the recorded run or are mocked, so critical side effects don't re-fire. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
AlexejPenner
changed the title
…raph It reads as hedging in this guide's intro; the carve-out stays on the compare pages where the canon mandates it. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
AlexejPenner
marked this pull request as ready for review
July 9, 2026 08:34
Contributor
Documentation Link Check Results✅ Absolute links check passed |
Contributor
|
@htahir1 heads-up, in-case you missed it. |
htahir1
approved these changes
Jul 13, 2026
Contributor
|
this time @AlexejPenner heads-up, in case you missed 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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Single-paragraph docs change: the agents-guide differentiation paragraph now follows the Kitaru messaging canon v3 — names Braintrust/LangSmith/Langfuse, states the watched-vs-executed distinction, notes that replay never re-fires yesterday's real actions, and keeps the single-shot honest boundary.
What changed
docs/book/user-guide/agents-guide/README.md: one paragraph replaced (1 insertion, 1 deletion). The §2 competitors LOCK and §6 boundary LOCK are embedded byte-identical to the canon. Hint box, frontmatter, and surrounding list untouched.What reviewers should focus on
Validation
Byte-level LOCK verification against the canon; adversarial fresh-eyes review against the rollout DONE checks — no failures; all relative links on the page verified on disk.
🤖 Generated with Claude Code
TODO (follow-up, not blocking review)
Backport the no-refiring sentence into the canonResolved in canon v3.2: the original claim was inaccurate (model calls do re-run live); the corrected replay-semantics sentence (tool calls answer from the recorded run or are mocked) is now a §2 lock and is used here.