talk-stage1-extract

Extracts and structures source material (articles, transcripts, notes) into a talk summary with narrative arc, themes, metrics, and gaps. Auto-detects REX vs Concept type. Use when starting a new talk from any source material or auditing existing material before committing to a talk.

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Advisory

Suggest reviewing before use

Quality

Content

57%

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

Well-organized and self-contained with a clear output template and signal table, but it reads more as a specification than an executable guide: steps are descriptive rather than runnable and lack inter-step validation feedback loops.

Suggestions

Tighten or merge the Anti-patterns and Tips sections, which overlap with the numbered steps and Validation Checklist, to reduce padding.

Add a validate-then-fix retry note after metric extraction and type detection (e.g. re-confirm detected type against the signal table before writing the summary) to create an explicit feedback loop.

Make at least one step executable — for instance, show the exact `AskUserQuestion` call for missing metadata or the literal file-write step — so the guidance is copy-paste ready rather than descriptive.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and avoids explaining concepts Claude already knows, but it is fairly long and somewhat padded — e.g. the Anti-patterns, Tips, and Related sections restate ideas already covered by the step list and validation checklist.	2 / 3
Actionability	It gives concrete structure (a full output template, detection-signal table, metric-format examples) but stops short of executable commands — steps like 'Reads the source' and 'Extracts the narrative arc' are descriptions rather than copy-paste-ready instructions, and no code/commands are provided.	2 / 3
Workflow Clarity	The eight numbered steps give a clear sequence and the Validation Checklist provides a checkpoint, but there is no validation feedback loop (validate -> fix -> retry) between steps — e.g. after type detection or metric extraction there is no re-check before writing the file.	2 / 3
Progressive Disclosure	No bundle files exist and the skill is appropriately self-contained: a single SKILL.md with well-organized sections (When to Use, What It Does, Input, Output, Output Format) and one-level-deep Related links to sibling stages — clean, no nested references.	3 / 3
	Total	9 / 12 Passed

Description

77%

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

A strong description: concrete actions, an explicit 'Use when' trigger, and a clearly distinct talk-pipeline niche. Its weakness is trigger-term naturalness — it leans on internal jargon (REX/Concept, narrative arc) over phrases users would actually say.

Suggestions

Add user-natural trigger terms such as 'summarize a talk', 'prepare a presentation from source material', or 'audit research before writing a talk' alongside the existing 'starting a new talk' phrasing.

Soften jargon like 'REX vs Concept' and 'narrative arc' in the description, or pair each with a plain-language equivalent so the trigger reads naturally to non-pipeline users.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Extracts and structures source material', 'summary with narrative arc, themes, metrics, and gaps', and 'Auto-detects REX vs Concept type' — which match the anchor for listing several specific concrete actions.	3 / 3
Completeness	It answers both what ('Extracts and structures source material...') and when ('Use when starting a new talk from any source material or auditing existing material before committing to a talk') with an explicit 'Use when' clause.	3 / 3
Trigger Term Quality	'articles, transcripts, notes' and 'starting a new talk from any source material' are relevant but it misses natural user phrasings like 'summarize my talk', 'prepare a presentation', or 'audit source material'; 'narrative arc' and 'REX/Concept' are jargon rather than terms a user would say.	2 / 3
Distinctiveness Conflict Risk	The talk-pipeline niche is clear and the trigger ties to a specific pipeline stage ('starting a new talk', 'before committing to a talk'), making it unlikely to fire for unrelated skills.	3 / 3
	Total	11 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 3 suspicious	Warning

	Total	14 / 16 Passed

Repository: FlorianBruniaux/claude-code-ultimate-guide
Commit: eb4fbbd

Reviewed: 2 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.