docs(openspec): phase6 tracker cleanup — check off done T7 tasks + corrections

[chrisjz]

What

Tracker hygiene pass on phase6-tracking/tasks.md after the C3-assembly / metric / re-measure PRs (#249/#250/#251), bringing the T7 checkbox state and notes in line with what's actually done.

Changes

Check off completed-but-unmarked tasks:

• T7.prep.connectome_continuous (Stage 2) — connectome climbed C1→C2→C3, integrated config committed, re-measured ~52% mid-pack (added a Stage-2-complete summary sub-bullet).
• T7.prep.continuous_behaviours (parent) + s3_generic_and_assemble — LSTM/CfC behaviours + the 6 assembled C3 configs.

Add what was missing:

• T7.prep.convergence_metric — the level-agnostic ranked-metric fix (feat(benchmark): level-agnostic plateau convergence detection #250) had no tracker entry despite being a hard prerequisite for the ranking. Added marked-done with the why.
• A Stage-6 readiness note — configs committed + recipes frozen, the 3000ep budget, the level-agnostic metric, and the revised converged ranking pending n≥8.

Clarity + correctness:

• Annotated T7.prep.should_may_continuous as non-gating for the MUST ranking (it gates only optional SHOULD/MAY arm inclusion, deferred to after Stage 6) so it doesn't read as a forgotten prerequisite.
• Fixed a self-contradictory ranking line (branch review): it listed connectome 52 > LSTM (~59 tuned), placing tuned-LSTM last despite 59 > 52. Corrected in both tracker occurrences and the 5 config headers to state the 2400ep baseline order (LSTM 45 last) and that the re-tune lifts LSTM to ~59 into the CfC/connectome/LSTM mid-cluster (~52–59), internal order pending the uniform-budget n≥8.

Audit result

Every prerequisite for the MUST n≥8 ranking is now checked off; the prep section is complete for the 6 MUST arms. Remaining [ ] tasks are genuinely outstanding (the n≥8 cells themselves, ablations, SHOULD/MAY arms, separation experiments, real-worm validation, analysis, logbook) — nothing blocks the final runs.

Validation

openspec validate phase6-tracking --strict ✓; all 5 configs parse; pre-commit (yaml + mdformat + markdownlint) ✓; no tmp/ or home-path leaks.

🤖 Generated with Claude Code

Summary by CodeRabbit

• Documentation
  • Updated performance metrics documentation across multiple training scenario configurations, incorporating revised benchmark rankings and refined baseline performance assessments
  • Refreshed Phase 6 task tracker to reflect stage completion milestones, finalized configuration parameters, and updated development readiness status for integrated training initiatives

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: b9d4e407-e5d0-4aa9-87d3-16e7da959e29

📥 Commits

Reviewing files that changed from the base of the PR and between 948d575 and 2a31bf3.

📒 Files selected for processing (6)

• configs/scenarios/foraging_predator_thermal/cfcppo_small_continuous2d_combined_klinotaxis.yml
• configs/scenarios/foraging_predator_thermal/connectomeppo_small_continuous2d_combined_klinotaxis.yml
• configs/scenarios/foraging_predator_thermal/lstmppo_small_continuous2d_combined_klinotaxis.yml
• configs/scenarios/foraging_predator_thermal/mlpppo_small_continuous2d_combined_klinotaxis.yml
• configs/scenarios/foraging_predator_thermal/transformerppo_small_continuous2d_combined_klinotaxis.yml
• openspec/changes/phase6-tracking/tasks.md

---

📝 Walkthrough

Walkthrough

Inline ranking comments in five foraging_predator_thermal integrated-C3 scenario YAML configs are updated to reflect a revised mid-cluster ordering with LSTM's re-tune lift and unsettled internal order. The Phase 6 T7 task tracker marks prep tasks complete, records locked parameters, and adds a Stage-6 readiness block.

Changes

Phase 6 T7 Ranking Comment and Tracker Sync

Layer / File(s)	Summary
Plateau metric ranking comments in scenario configs configs/scenarios/foraging_predator_thermal/mlpppo_small_continuous2d_combined_klinotaxis.yml, configs/scenarios/foraging_predator_thermal/transformerppo_small_continuous2d_combined_klinotaxis.yml, configs/scenarios/foraging_predator_thermal/cfcppo_small_continuous2d_combined_klinotaxis.yml, configs/scenarios/foraging_predator_thermal/connectomeppo_small_continuous2d_combined_klinotaxis.yml, configs/scenarios/foraging_predator_thermal/lstmppo_small_continuous2d_combined_klinotaxis.yml	Comment blocks at lines 13–15 in all five scenario configs are revised from a simplified static ranking to a mid-cluster description noting LSTM's score lift from ~45 to ~59 via re-tune and that the internal CfC/connectome/LSTM order is unsettled pending n≥8 runs.
Phase 6 T7 task tracker openspec/changes/phase6-tracking/tasks.md	T7.prep.connectome_continuous and T7.prep.continuous_behaviours are marked complete with locked difficulty/recipe parameters recorded. The level-agnostic convergence metric entry is updated to note the sub-saturation plateau ranking-corruption fix. A new Stage-6 readiness block documents frozen integrated-C3 configs, 3000ep budget, and the revised ranking with n≥8 statistics pending.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

Possibly related PRs

• SyntheticBrains/nematode#251: Updates the same foraging_predator_thermal scenario header ranking comments and Phase 6 T7 tracker notes covering the identical LSTM retune (45→59) narrative and Gate 3 ordering revision.
• SyntheticBrains/nematode#249: Updates per-architecture level-agnostic plateau ranking outcomes in the same scenario comment blocks and openspec/changes/phase6-tracking/tasks.md.
• SyntheticBrains/nematode#229: Revises the same Phase 6 tracking file around the level-agnostic convergence metric and enforces the n≥8 statistical-power constraints that this PR references.

Poem

🐇 Hop, hop! The LSTM climbed the hill,
From forty-five to fifty-nine — what a thrill!
The mid-cluster order still flutters like a leaf,
Eight seeds must fall before we find relief.
The configs are frozen, the budget is set,
The convergence detector catches our net! 🌿

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately reflects the main content: cleanup of phase 6 tracking documentation, checking off completed T7 tasks, and making corrections to metrics and ordering.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch openspec/phase6-tracker-cleanup

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!