Repository navigation
π proposal: Nous β Spek β the winning experiment as a spek-new source - #17106
Conversation
Design-only straw man for the Nous team conversation #17020 asks for. Records what the external Nous framework leaves on disk for a winning candidate (best_found.json, meta_findings.json deployment_recommendation, findings, bundle code_changes), what Hive reads of it (ledger.json and principles.json at boot, in a vocabulary the framework's own ledger rows do not share), and what spek-new accepts (one sources[].uri). Proposes an artifact shape, publishing it as a GitHub issue so spek-new can start from the link with no Hive or Spek change, and lists the open questions. Refs #17020 Hive-Run: #17020 Hive-Plan: #17020 proposal Hive-Spec: nous-spek-source#straw-man Signed-off-by: Karibhou <Karibhou@users.noreply.github.com>
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
Open PR overlap checkBase branch: Conflicts are checked with every PR merged onto the current tip of
|
There was a problem hiding this comment.
Decision needed: @clubanderson should decide whether merging this design-only straw man should close #17020 (the PR body says Closes #17020, and v5 is the default branch so it will), or whether the body should carry only Refs #17020 as the original commit does, keeping the issue open for the Nous-team questions the document itself lists as unanswered.
The content is fine; the issue link is the one thing a person has to settle before this merges.
The mismatch
- PR body:
Closes #17020andRefs #17020on consecutive lines. Only the first one has an effect βv5is the repository default branch, so GitHub will close #17020 when this merges. - Original commit
edb20e4(@Karibhou, preserved in this adoption) saysRefs #17020only, and its message calls the doc a "design-only straw man for the Nous team conversation #17020 asks for". src/docs/design/nous-spek-source.md:3β "Status: proposed β design only, no Hive change.";:129β "## Open questions for the Nous team";:147β "## What Hive would change, once agreed".- #17020 itself: "Placeholder for a direction, not a spec yet β¦ No Hive change is proposed yet; this issue exists so the direction has a home and can be linked from the Nous work."
So merging as-is closes the direction's "home" while the direction is still open. If that is intended (the straw man is the home now), fine β but then the stray Refs line should go. If not, drop Closes and keep Refs #17020 with a note on what remains (the Β§Open questions).
What I checked in the doc
bin/nous-install.shexists at this head as the doc says, and theledger.json/principles.jsonboot-read it describes (:60-63) matches whatsrc/pkg/convergencereferences; I did not independently verify the Nous-side file shapes (best_found.json,meta_findings.json) against the pinned external commit.src/docs/design/README.md:201-209adds the index entry in the same form as its neighbours, with the status marker up front.
No executable change; nothing else blocking from me.
β hive: agent=reviewer-2 backend=copilot model=claude-fable-5.1 copilot=1.0.88
|
Hive status β updated in place; edits do not notify. Full history is in the Activity Monitor. hive/awaiting-ci: retitled to π Hive Agent: β hive: agent=scanner backend=copilot model=claude-fable-5.1 copilot=1.0.88 Review findings from Read the page against the tree at
The README index entry matches the page's status line. Review findings from Read at head The one open ask from #17063 was carried over unresolved, and the adoption made it worse. This body now contains both GitHub honours the Fix is body-only, no push: delete the Everything else looks correct to me as a proposal document: design-only, no Hive behaviour change, README index entry follows the existing |
There was a problem hiding this comment.
Read both files. The design doc's code references check out on v5: loadNousStateFromPaths is at src/cmd/hive/main.go:6872, the id fallback chain id/experiment_id/experimentId/name is at src/pkg/dashboard/api_nous_progress.go:242, and src/pkg/nous exists. Its relative links resolve: src/docs/spektacular.md:84 (## Publish hand-off) and src/docs/strategy-lab.md:281 (### The experiment loop), plus the new index entry in src/docs/design/README.md. Content looks fine to me.
The red checks are not caused by the diff:
PR Type: this repo accepts only β β¨ π π π π± as title prefixes;πis not one of them. Retitle toπ docs: proposal β Nous β Spek β¦(keeps the same meaning and passes the check).Docs Link Check: the only failure isCHANGELOG.md: broken link 'url', pre-existing onv5and fixed by #17096; it clears on rebase after that merges.
One body nit: it has both Closes #17020 and Refs #17020 β keep just Closes #17020 if this design doc is what #17020 asked for.
β hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88
|
Thank you for your contribution! Your PR has been merged. We'd love to hear how your experience was: share feedback |
Supersedes #17063 (adopted after CI went red and the contributor was unavailable; original commits by @Karibhou preserved).
Refs #17020 β does not close it. The page is a straw man for the Nous conversation that issue asks for; #17020 stays open until the Nous team answers the "Open questions for the Nous team" section (winning-candidate definition / Findings Review gate, who assembles the artifact, patch-vs-intents, where the artifact is published, the campaignβrun join id, and who writes the Hive-shaped ledger keys). Once agreed, a follow-up PR will carry the Hive change and close the issue.
This branch is #17063's head
edb20e47ddc1as pushed by @Karibhou (forkKaribhou/hive), unchanged. Last lane reason on the original:contributor-lane:changes-requested.Hive fix agent: rebase this branch on
v5, fix the failing checks / review asks, and push to THIS branch. Keep the original author's commits and theirSigned-off-by; add your own sign-off on your commits. Do not open another PR.