Repository navigation
Make Forge briefing eval publish a full brief - #103
Draft
tarun-etikala wants to merge 2 commits into
Draft
tarun-etikala wants to merge 2 commits into
tarun-etikala wants to merge 2 commits into
Conversation
tarun-etikala
changed the title
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The Forge morning-briefing eval could complete with a text response without publishing the canonical full
brief.json. The sandbox lacked a synthetic installationUSER.md, and the pipeline pinned an older SAW agent image that did not follow the current brief-reader/full-publication flow.This PR adds a fixture profile aligned with the mock M365 scene and passes it through the harness's
AGENT_EVAL_FORGE_USER_FILEhook when present. An explicit harness setting still takes precedence, and other submissions are unaffected. It also pins the verified SAW agent image in the OpenShell pipeline, Task, and example PipelineRun.Validation: 47 focused tests and pre-commit passed. Two clean-source integration runs in
forge-nommenusing Harness #9, Flow #101, this PR, and the OIDC refresh in #104 succeeded all five Tasks: run 1, run 2. Both published ascope: fullbrief.json, passedpublished_briefat 100%, and scored 0.70 against the 0.50 threshold. These validate pipeline execution and publication; the briefing content still has low recall scores.Companion changes: Harness #9, Flow #101, and Flow #104.