Add OrcaReplay to Observation - #540
Open
xizhuomengcontin wants to merge 1 commit into
Open
Conversation
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Author
|
/kind documentation |
Author
|
/kind documentation |
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds OrcaReplay to
### Observation, in alphabetical position between OpenLLMetry and wandb. One line, entry format and badge set match the surrounding rows.What it is. A record-replay debugger for AI agents. It records a run at the process and socket boundary — a local proxy in front of the model API plus a PATH shim for exit codes, a JSON-RPC tee for MCP, a shadow git index for per-turn file changes — then replays that run offline byte-for-byte, or forks it from any checkpoint onto a different model.
Why Observation rather than Tool. The neighbours in this section (Latitude, OpenLLMetry, wandb) answer what happened and what it cost. OrcaReplay covers the same ground and adds re-execution: the recorded trace runs again with the network off, so the failure can be stepped through repeatedly instead of read once. The trace format is a normative spec published under CC BY 4.0, so other readers can reimplement it — there is already a Python one alongside the TypeScript core.
Apache-2.0, Node 20+, 1,393 tests, published on npm as
orcareplay. Also ships an MCP server (orca mcp) so an agent can read and replay its own runs.Disclosure: I work on OrcaReplay.
🤖 Generated with Claude Code