architecture-griller: add second griller (advisory-only structural-fit)#31
Conversation
The second griller (role: griller) at pharn-pipeline/grillers/architecture/, mirroring the testability griller (#29). Interrogates a PLAN along the architecture axis: does the approach fit the existing tree, or introduce layering / sibling-coupling (P3) inconsistency? Honest floor/advisory split (OQ1): griller MEMBERSHIP is the only runtime floor (count-grillers.mjs, reused unchanged); the architectural-fit assessment is entirely ADVISORY. No manufactured floor sub-check — a genuine deterministic invariant (pharn-contracts purity) belongs in validate.mjs, not an advisory griller. enforces: ["P3"] is eval-bound by the misfit fixture; the trust-fence (needle_absent) is intact. validate GREEN (3 capabilities); regress no-regressions; verify PASS. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Malformed nested back-ticks in the L-DEL-1 candidate lesson tripped markdownlint MD038, leaving the whole-repo lint:md gate red (blocking every later feature's verify). Rewritten to render DELETE as bold + a plain path code span — no change to the lesson's meaning. Surfaced by the architecture-griller verify run; fixed separately. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
|
Warning Review limit reached
Next review available in: 47 minutes Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (17)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
…ed verifies) L11 — verify's whole-repo format:check/lint:md (added by L9) run once at HEAD with no base comparison, so a pre-existing style error in an unrelated committed file blocks every later feature's verify (unlike /pharn-dev-regress, which classifies base-red as pre-existing). Surfaced by this feature's verify run and human-approved via gated /pharn-dev-memory-promote; the sibling candidate L12 was denied as covered by L5. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
1 participant
What
The second griller — a
role: grillercapability at ROOTpharn-pipeline/grillers/architecture/, mirroring the testability griller (#29). It interrogates a PLAN along the architecture axis: does the approach fit the existing tree, or introduce structural inconsistency (layering violations, sibling coupling P3 forbids, new patterns where established ones exist)?The honest floor/advisory split (the point of this increment)
Architecture-fit is irreducible judgment, so this griller proves the converse #29 anticipated: a griller may be advisory-only beyond membership.
role: griller, counted bycount-grillers.mjs, reused unchanged →registered1→2;validate2→3 capabilities).pharn-contractspurity) is explicitly routed tovalidate.mjs(the floor over built product), not bolted onto an advisory griller.enforces: ["P3"]is eval-bound by the misfit fixture; the trust-fence (needle_absent_from_enum_gated) is intact. "Ensures good architecture" is struck as the disease (P0).Gates (this run)
validateGREEN (3 capabilities) · regress no-regressions · verify PASS (test 167 / validate / lint / format:check / lint:md all 0) · review GREEN (0 blocking findings).Two commits
REVIEW.mdthat was leaving the whole-repolint:mdgate red (blocking every later feature's verify). Surfaced by this feature's verify run; unrelated to the griller; split into its own commit for a clean review.Proposed lessons (candidates in REVIEW.md — not yet canon;
/pharn-dev-memory-promotegates them)lint:mdblocking an unrelated feature's verify.package.jsontest glob, not a hand-expanded file list.🤖 Generated with Claude Code
Summary by CodeRabbit
New Features
Bug Fixes
Documentation