Skip to content

feat: generate shared standards discovery across registered repos #474

Description

@Jammy2211

Generate and distribute shared standards discovery

Type: feature
Target: PyAutoMind
Difficulty: medium
Autonomy: supervised
Consequence: judge
Review-minutes: 10
Unattended: needs-slicing
Status: draft

Original user request (verbatim)

ok do those next tasks

Approved scope and plan

This is the approved instruction-discovery phase of draft/feature/pyautobrain/standardize_dashboard_orchestration_prompt_panel.md. Reuse Brain docs/standards.md as canonical contract and preserve all existing guidance.

  1. Add one concise Mind policy source, canonical GitHub standards link and explicit manifest board-owner metadata for thirteen board owners. Universal text asks agents to consult applicable standards on demand, identify consumers and validate adoption; only board owners get board component guidance.
  2. Extend scripts/repos_sync.py with a bounded standards writer/checker and repeatable --repo filter. Validate flags, identities and marker integrity before mutation; scoped standards writes must not touch any unrelated generated artifacts. Report missing/stale guidance and partial-checkout coverage honestly. Default universal audit remains strict for available registered checkouts; no permanent exceptions for temporarily claimed repos.
  3. Cover idempotence, preservation, missing/malformed/stale blocks, owner/nonowner variants, invalid selections with zero writes, and partial grouped/flat checkouts using meaningful synthetic fixtures. CI checks the generator and target-owned block without requiring not-yet-merged consumer mains to have migrated; document that scope explicitly.
  4. Generate blocks into all eligible repository worktrees. There are 46 registered repositories. PyAutoArray and euclid_strong_lens_modeling_pipeline are deferred because of active claims. Nine panel consumers receive generated guidance on their existing panel task branches; Brain follows after sizing fix: repos_sync check mode never checks root AGENTS.md routing table / WORKFLOW owner map #480 clears. All others get isolated standards-discovery worktrees and instruction-only PRs. Preserve unrelated dirty canonical checkouts.
  5. Record coverage and exact generated diffs. Run generator tests and relevant drift checks, open tested PRs for source plus eligible destinations, and leave human /prm. Deferred destinations remain recorded work, not completed rollout. No RTD source move.

Branch survey

All 46 registered repositories exist on main with tracked clean AGENTS.md. Some have unrelated canonical changes; use isolated task branches. Initial claims: 34 repositories excluding the nine panel consumers, Brain sizing, and two external active claims. Branch feature/standards-discovery.

Tier: judge; merge mode: human /prm.

Activity

  1. Jammy2211 commented on Oct 6, 2026

    @Jammy2211
    CollaboratorAuthor

    PRs opened — 2026-10-06

    35 tested PRs open, plus nine generated consumer blocks carried by orchestration-panel-consumers. Generator 652 tests, 33 adversarial/standards cases, independent re-review CLEAN; strict Brain docs build passed. Combined source audit: 44/46 blocks present. Array and Euclid pipeline deferred for nnls-memo-scattered-backoff / vis-lp-inspection-bundle claims. Human /prm; do not close entire parent initiative until deferred destinations and publication verification are complete.

    Repository PR
    PyAutoCortex PyAutoLabs/PyAutoCortex#57
    PyAutoHeart PyAutoLabs/PyAutoHeart#287
    PyAutoFit PyAutoLabs/PyAutoFit#1660
    PyAutoGalaxy PyAutoLabs/PyAutoGalaxy#648
    PyAutoLens PyAutoLabs/PyAutoLens#769
    PyAutoReduce PyAutoLabs/PyAutoReduce#80
    PyAutoCTI PyAutoLabs/PyAutoCTI#113
    autofit_workspace PyAutoLabs/autofit_workspace#166
    autogalaxy_workspace PyAutoLabs/autogalaxy_workspace#255
    autolens_workspace PyAutoLabs/autolens_workspace#584
    autocti_workspace PyAutoLabs/autocti_workspace#36
    autoreduce_workspace PyAutoLabs/autoreduce_workspace#4
    autofit_workspace_test PyAutoLabs/autofit_workspace_test#106
    autogalaxy_workspace_test PyAutoLabs/autogalaxy_workspace_test#127
    autolens_workspace_test PyAutoLabs/autolens_workspace_test#344
    autocti_workspace_test PyAutoLabs/autocti_workspace_test#23
    autofit_workspace_developer PyAutoLabs/autofit_workspace_developer#28
    autolens_workspace_developer PyAutoLabs/autolens_workspace_developer#146
    HowToFit PyAutoLabs/HowToFit#70
    HowToGalaxy PyAutoLabs/HowToGalaxy#84
    HowToLens PyAutoLabs/HowToLens#95
    autocti_assistant PyAutoLabs/autocti_assistant#35
    autofit_assistant PyAutoLabs/autofit_assistant#54
    autogalaxy_assistant PyAutoLabs/autogalaxy_assistant#33
    autolens_assistant PyAutoLabs/autolens_assistant#152
    euclid_assistant https://github.com/Jammy2211/euclid_assistant/pull/14
    autolens_profiling PyAutoLabs/autolens_profiling#386
    autolens_inference PyAutoLabs/autolens_inference#19
    autolens_visualization PyAutoLabs/autolens_visualization#2
    autogalaxy_visualization PyAutoLabs/autogalaxy_visualization#2
    autofit_visualization PyAutoLabs/autofit_visualization#2
    autocti_visualization PyAutoLabs/autocti_visualization#2
    pyautolabs.github.io PyAutoLabs/pyautolabs.github.io#29
    PyAutoMind #475
    PyAutoBrain PyAutoLabs/PyAutoBrain#483

    CI was judged once at each exact head: no failures observed; pending, skipped and absent checks are not green. The corresponding CI snapshots and local validation artifacts are preserved in the session scratch evidence. No timers, watchers or automatic merge subscriptions remain.

  2. Jammy2211 commented on Oct 6, 2026

    @Jammy2211
    CollaboratorAuthor

    Progress: 44/46 merged (2026-10-06). Issue stays open for two deferred destinations

    Merged via /prm on 2026-10-06 (each head proven an ancestor of origin/main, all PRs merged=true):

    • Generator: feat: generate scoped shared-standards discovery #475 (7208711). Contract docs: docs: connect panel adoption and generated standards guidance PyAutoBrain#483 (bdebd4e1)
    • 33 consumer PRs: PyAutoCortex#57, PyAutoHeart#287, PyAutoFit#1660, PyAutoGalaxy#648, PyAutoLens#769, PyAutoReduce#80, PyAutoCTI#113, autofit_workspace#166, autogalaxy_workspace#255, autolens_workspace#584, autocti_workspace#36, autoreduce_workspace#4, autofit_workspace_test#106, autogalaxy_workspace_test#127, autolens_workspace_test#344, autocti_workspace_test#23, autofit_workspace_developer#28, autolens_workspace_developer#146, HowToFit#70, HowToGalaxy#84, HowToLens#95, autocti_assistant#35, autofit_assistant#54, autogalaxy_assistant#33, autolens_assistant#152, Jammy2211/euclid_assistant#14, autolens_profiling#386, autolens_inference#19, autolens_visualization#2, autogalaxy_visualization#2, autofit_visualization#2, autocti_visualization#2, pyautolabs.github.io#29
    • The nine panel consumers' generated blocks shipped on their own branches (feat: adopt shared orchestration panels across remaining boards PyAutoEars#15, now closed)

    That puts the generated standards block on main in 44 of 46 registered destinations.

    CI and live checks: CI was judged once at each exact head. The skipped legs were all expected skips: Heart's smoke relevance gate on AGENTS.md-only diffs, Reduce unittest-nojax excluded for autoreduce, and Mind Spawn Drift running on push only. Cortex, both *_workspace_developer repos and pyautolabs.github.io have no applicable CI, and each passed a local repos_sync.py --check --only "shared-standards blocks (generated)". RTD standards.html and board-orchestration.html went live at 14:40:08 UTC. All nine panel boards were checked live at about 14:58 UTC.

    Deferred (not shipped):

    Follow-up: PyAutoMind/draft/feature/pyautomind/standards_discovery_deferred_destinations.md. It is blocked by both claims. Once each claim clears, it runs the bounded repos_sync.py --write --only "shared-standards blocks (generated)" --repo NAME for that repo and validates preservation and drift. It then restores broad firewall standards checking, runs the universal completion audit, and closes this issue.

    Record: PyAutoMind/complete/2026/10/standards-discovery.md (Mind 3f1f7f3).

  3. Jammy2211 commented on Oct 6, 2026

    @Jammy2211
    CollaboratorAuthor

    Shipped — initiative complete (2026-10-06)

    The generated shared-standards discovery block is now on main in 46 of 46 registered destinations, and the firewall checks it universally again.

    Final wave (merged 2026-10-06 via /prm; each head proven an ancestor of origin/main):

    Universal completion audit: I ran repos_sync.py against fresh clones of all 46 registered origin/mains. --check --only "shared-standards blocks (generated)" reported OK on 46/46, with no deferred targets and no missing clones. The full --check passed on every leg.

    Whole initiative:

    Records:

    • PyAutoMind/complete/2026/10/standards-discovery.md (44/46 wave)
    • PyAutoMind/complete/2026/10/orchestration-panel-consumers.md (panel adoption and live verification)
    • PyAutoMind/complete/2026/10/standards-discovery-deferred.md (this final wave)

    The parent initiative prompt, standardize_dashboard_orchestration_prompt_panel.md, is now fully shipped and has its own completion record. Closing this issue.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions