Transform CHANGELOG entries into social content (LinkedIn, Twitter/X, Newsletter, Slack) in FR + EN. Use after releases or weekly to generate ready-to-post content from guide updates.
Install with Tessl CLI
npx tessl i github:FlorianBruniaux/claude-code-ultimate-guide --skill guide-recap93
Quality
92%
Does it follow best practices?
Impact
94%
1.17xAverage score across 3 eval scenarios
Tone and anti-hype content rules
No hype words (EN)
100%
100%
No FOMO/urgency (EN)
100%
100%
No invented metrics
100%
100%
Concrete CHANGELOG numbers used
100%
100%
LinkedIn emoji budget
100%
100%
Twitter emoji budget
100%
100%
No forbidden emojis
100%
100%
LinkedIn links to landing
100%
100%
Twitter links to GitHub
100%
100%
Source credited
100%
100%
FR vouvoiement in LinkedIn
100%
100%
No clickbait hooks
100%
100%
Without context: $0.1396 · 41s · 8 turns · 13 in / 2,673 out tokens
With context: $0.3610 · 1m 22s · 14 turns · 1,778 in / 4,433 out tokens
CHANGELOG parsing, scoring, and entry selection
NEW_CONTENT category assigned
55%
100%
GROWTH_METRIC category assigned
62%
100%
MAINTENANCE entries excluded
55%
100%
Scoring formula applied
22%
100%
Highest-scored entry is hook
100%
100%
Jude Gao credited
100%
100%
reference.yaml skipped
25%
100%
Concrete numbers in posts
100%
100%
Version framing correct
100%
100%
Scoring report present
100%
100%
8 posts generated
100%
100%
FIX entry handling
100%
100%
Without context: $0.2492 · 1m 19s · 11 turns · 55 in / 4,785 out tokens
With context: $0.8656 · 2m 58s · 28 turns · 5,147 in / 11,825 out tokens
Format-specific constraints and link targets
Week framing in headers
88%
100%
Release count prefix
100%
100%
Maintenance entries excluded
22%
22%
Shen & Tamkin credited
100%
100%
LinkedIn links to landing page
62%
100%
Twitter links to GitHub
0%
100%
Slack links to GitHub
0%
100%
LinkedIn FR uses vouvoiement
100%
50%
Twitter FR uses tutoiement
37%
37%
Slack has no hashtags
100%
100%
Output saved to correct path
100%
100%
Newsletter has both links
100%
100%
LinkedIn fixed hashtags present
25%
100%
Without context: $0.2779 · 1m 9s · 15 turns · 19 in / 4,662 out tokens
With context: $0.7468 · 2m 49s · 23 turns · 5,270 in / 11,294 out tokens
Table of Contents
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.