extract-transcripts

Extract readable transcripts from Claude Code and Codex CLI session JSONL files

11.87x

Quality

—

Does it follow best practices?

Impact

95%

11.87x

Average score across 3 eval scenarios

Securityby

Risky

Do not use without reviewing

Quality

Content

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a tightly organized, highly actionable CLI reference with executable commands, but it could be leaner by factoring repeated invocations, and its progressive disclosure is weakened by inline detail and referenced scripts that are missing from the bundle.

Suggestions

Define each script invocation once (e.g. via a shell variable or a short alias) and reference it in subsequent examples to remove the repeated full-path verbosity.

Split the long DuckDB indexer command catalog into a separate reference file linked from the overview, so the main SKILL.md stays a concise entry point.

Either bundle the referenced scripts under scripts/ or note explicitly where they are obtained, since the paths currently point to files absent from the skill bundle.

Dimension	Reasoning	Score
Conciseness	The body avoids explaining concepts Claude already knows and stays command-focused, but it repeats the full 'uv run ~/.claude/skills/extract-transcripts/extract_transcript.py' path roughly 14 times instead of defining the invocation once and referring back to it.	2 / 3
Actionability	Every example is a concrete, copy-paste-ready bash command with real flags and file paths, accompanied by an Options list explaining each flag, matching the 'fully executable, copy-paste ready' anchor.	3 / 3
Workflow Clarity	Each task is a single unambiguous command and the operations are read-only (no destructive or validation-needing batch steps), so the simple-skill carve-out applies and the well-grouped command sections are clear.	3 / 3
Progressive Disclosure	The body is organized into clear sections, but all detail is inline in one ~95-line file with no one-level-deep reference files, and the scripts it references by path (extract_transcript.py, extract_codex_transcript.py, transcript_index.py) are not present in the bundle.	2 / 3
	Total	10 / 12 Passed

Description

57%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific and clearly differentiated, naming a concrete action and a distinct niche, but it lacks an explicit 'Use when...' trigger clause and only partial natural-keyword coverage.

Suggestions

Add an explicit trigger clause, e.g. 'Use when the user wants readable transcripts or exports of Claude Code or Codex CLI chat/session history.'

Broaden trigger terms to include natural variations users might say, such as 'chat history', 'conversation logs', or 'export session'.

Optionally list a second concrete action (e.g. 'search across sessions') to raise specificity from a single action to multiple.

Dimension	Reasoning	Score
Specificity	The phrase 'Extract readable transcripts from Claude Code and Codex CLI session JSONL files' names a concrete action and a specific domain, but lists only a single action rather than multiple specific concrete actions as required for a 3.	2 / 3
Completeness	It clearly answers 'what' (extract readable transcripts from session JSONL files) but provides no 'Use when...' trigger guidance, which per the judging guidelines caps completeness at 2.	2 / 3
Trigger Term Quality	It includes relevant natural terms a user would say ('transcripts', 'Claude Code', 'Codex CLI', 'session') but misses common variations such as 'chat', 'conversation', 'history', or 'export', so coverage is partial.	2 / 3
Distinctiveness Conflict Risk	The combination of two named CLI tools plus 'session JSONL files' and 'transcripts' carves a clear niche that is unlikely to trigger for unrelated skills.	3 / 3
	Total	9 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: NeverSight/skills_feed
Commit: ff6a653

Reviewed: 5 days ago

Table of Contents

Discovery Implementation Validation

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.