extract-transcripts
Extract readable transcripts from Claude Code and Codex CLI session JSONL files
75
Quality
63%
Does it follow best practices?
Impact
95%
11.87xAverage score across 3 eval scenarios
Risky
Do not use without reviewing
Optimize this skill with Tessl
npx tessl skill review --optimize ./data/skills-md/0xbigboss/claude-code/extract-transcripts/SKILL.mdEvaluation results
100%
84%Transcript Extraction Helper
Single session extraction with optional flags
Correct script path
0%
100%
Uses uv run
0%
100%
--include-tools flag used
0%
100%
--include-thinking flag used
0%
100%
--summary flag used
0%
100%
Output to file with -o
0%
100%
Output file named transcript.md
0%
100%
README explains --include-tools
40%
100%
README explains --include-thinking
40%
100%
README explains --summary
40%
100%
Session file path used
100%
100%
No wrong interpreter
100%
100%
Without context: $0.3826 · 2m 5s · 18 turns · 22 in / 7,602 out tokens
With context: $0.3977 · 1m 22s · 22 turns · 26 in / 4,380 out tokens
85%
75%Bulk Session Archive for Code Review
Batch directory processing with skip-empty filtering
Correct script path
0%
100%
Uses uv run
0%
100%
--all flag used
0%
0%
--skip-empty flag used
0%
100%
--min-messages set to 3
0%
100%
Output to transcripts directory
0%
100%
Correct source directory
100%
100%
plan.md mentions flags
0%
100%
Without context: $0.4814 · 2m 16s · 21 turns · 24 in / 7,995 out tokens
With context: $0.3990 · 1m 25s · 22 turns · 276 in / 4,786 out tokens
100%
100%Session Search Workflow for an AI-Assisted Development Blog
DuckDB transcript index build and search workflow
Correct index script path
0%
100%
Uses uv run
0%
100%
index subcommand used
0%
100%
--full reindex flag
0%
100%
recent subcommand used
0%
100%
--limit flag on recent
0%
100%
--project flag on recent
0%
100%
--since flag on recent
0%
100%
search subcommand used
0%
100%
--cwd flag on search
0%
100%
show subcommand used
0%
100%
--summary flag on show
0%
100%
Database location mentioned
0%
100%
Without context: $0.3151 · 1m 51s · 17 turns · 19 in / 6,209 out tokens
With context: $0.3753 · 1m 16s · 21 turns · 26 in / 4,366 out tokens
- Repository
- NeverSight/skills_feed
- Commit
5342bca
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
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.