8a3e109b35
fix(linkedin-studio): S7 harden batch — format-aware length band on per-post quality check
...
S7 close-out (1 edit / 1 command-file · lint 81/0 · counts 29/19/26/6):
- batch.md:95 Step-3b Quick Quality Check: replace vague "Character count in range" with a format-aware band (standard 1,200-1,800 · quick 150-500 · carousel/video caption per that format) — the old line named no band, and the 3a structure (:88-92) sums to 960-1640, under the 1,200 standard floor, so a ~1,000-char standard post passed "in range" unflagged
Axis-c (length band) closed; axes a/b/d PASS. Class: hybrid guided/stateful + post-emitting per post (each of 3-5 drafts is a feed post). Buzzword (:98) + no-body-link (:97) already present — the gap S2-S6 found is already closed in batch. Wiring verified: trend-spotter + content-planner resolve; all 5 node helpers (queueUpcoming/queueAdd 8-arg/queueFormatSummary/generateIcalFromQueue/writeIcalFile) export with matching signatures. 3 deferrals (3a single-structure vs mixed-format promise · no per-post differentiation gate by design · weekly_goal-unset micro-fallback). Local-only per push policy.
2026-06-18 15:29:31 +02:00
1524b8438f
fix(linkedin-studio): S6 harden video — caption buzzword gate + voice-samples graceful degradation
...
S6 close-out (2 edits / 1 file · lint 81/0 · counts 29/19/26/6):
- video.md:129 Step-5 De-AI gate: add content rule #4 buzzword strip to the post-caption gate (only buzzword line was :106 under "Natural Speech Test" = script not caption; caption is post-emitting feed text)
- video.md:42/:45 Step-0: soften "(REQUIRED before scripting)" → "(read before scripting)" + add fresh-adopter graceful-degradation fallback (empty/absent voice-samples → SKILL.md/defaults, no dead-end)
Axis-c (buzzword on caption) + axis-d (voice-samples dead-end) closed; axes a/b PASS. Class: hybrid guided/stateful + post-emitting on the caption (hook 110-140 deliberately not imposed on the accompanying caption — operator-agreed). video-scripter + differentiation-checker wiring verified; 2.5-wps budgets exact. 2 deferrals (caption-hook class boundary + :69 "zero distribution" inter-reference honesty). Local-only per push policy.
2026-06-18 15:20:05 +02:00
d3efe326b4
fix(linkedin-studio): S5 harden carousel — Step-5 caption buzzword gate
...
S5 close-out (1 edit / 1 file · lint 81/0 · counts 29/19/26/6):
- carousel.md:123 Step 5 De-AI gate: add content rule #4 buzzword line to the caption gate (was missing; grep NONE pre-edit; caption is post-emitting feed text)
Axis-c (buzzword on the caption) closed; axes a/b/d PASS. Class: hybrid guided/stateful + post-emitting on the caption. mcp-image graceful degradation + differentiation-checker wiring verified. 2 deferrals (no-body-link on caption + cosmetic slide line-count). Local-only per push policy.
2026-06-18 14:28:07 +02:00
a3d26b223c
fix(linkedin-studio): S4 harden multiplatform — Step-0 empty-source fallback + cross-platform buzzword line
...
S4 close-out (2 edits / 1 command file · lint 81/0 · counts 29/19/26/6):
- multiplatform.md:28 Step 0: empty-source fallback (drafts empty + no recent posts + nothing pasted -> stop, ask user to paste; never fabricate source)
- multiplatform.md:45 Adaptation Templates: cross-platform voice-hygiene line (banned buzzwords = content rule #4 ; adaptation rewrites can introduce them)
Class call: NOT post-emitting (output is Twitter/slides/YouTube) -> no LinkedIn predicate imposed; tested as guided/cross-format + routing sub-predicate. Routing (/linkedin:newsletter) + both reference files resolve; no subagents. Axis-c (buzzword) + axis-d (degradation) closed; axes a/b PASS. Local-only per push policy.
2026-06-18 14:10:17 +02:00
b8c2155003
fix(linkedin-studio): S3 harden react — Step-6 buzzword gate + WebFetch-failure fallback (both paths)
...
S3 close-out (3 edits / 1 file · lint 81/0 · counts 29/19/26/6):
- react.md:131 Step 6: add no-buzzword checklist line (was missing; banned list = content rule #4 ; covers Step 6b via "Same as Step 6")
- react.md:60 Step 2: WebFetch-failure fallback (paywall/JS/YouTube-no-transcript -> ask user to paste or pick another; never fabricate)
- react.md:194 Step 2b: same fallback pointer for the multi-URL comparison path
Axis-c (buzzword) + axis-d (graceful degradation) closed on both paths; axes a/b PASS. Shared scorecard already clean post-S2. Local-only per push policy.
2026-06-18 13:44:00 +02:00
69840e7a94
fix(linkedin-studio): S2 harden post — Step-5 buzzword gate + scorecard hook-band consistency
...
P4 close-out (4 edits / 2 files · lint 81/0 · counts 29/19/26/6):
- post.md Step 5: add no-buzzword checklist line (was missing; banned list = content rule #4 )
- quality-scorecard.md l.32/114: one-sided "140" hook bound -> "110-140" (matches l.128, resolves self-contradiction)
- quality-scorecard.md: add no-buzzword line to Pre-Publish Basic Checklist
Shared scorecard fix benefits 8 consumers; only post hardened this session (react/batch/video/pipeline get own EVALUATE in S3/S5/S6/S8). Local-only per push policy.
2026-06-18 13:32:56 +02:00
94760cde90
fix(linkedin-studio): S1 hardening — method calibration (quick) + Start journey
...
Hardening phase S1. Tightened the hook character bound from one-sided
(under 140) to the full canonical 110-140 band across the Start-journey
creation surfaces, matching hooks/prompts/content-quality-gate.md:
- quick.md: hook band on Step 2 + Step 5 checklist; added a buzzword check
and a 150-500 length-band check; checklist tally 6 -> 7.
- onboarding.md: hook band in Phase 3.2 + 3.3 (length + no-links already present).
- first-post.md: hook band in Step 4 + Step 5 (length + no-links already present).
- setup.md: zero-edit pass (all four axes already satisfied).
Adds docs/hardening/log.md (per-command audit trail, 5-step method) and
docs/hardening/review.md (cold /trekreview: ALLOW, 0 BLOCKER/0 MAJOR/1 MINOR).
Lint Failed:0, counts 29/19/25/6 unchanged. No structural/version churn.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-05-31 06:46:11 +02:00
