roast

Use when someone asks to roast an idea, pressure-test or stress-test an idea, validate a business idea, "convene the panel", get a brutal second opinion before building something, or says "/roast". Spins up a 5-angle panel (Critic, Champion, Analyst, Investigator, Customer) that attacks the idea from every angle, then a Judge returns one GO / RESHAPE / KILL verdict with the cheapest test to de-risk it.

Quality

94%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

92%Weight 40%Scale 1-5

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

A well-engineered skill body: a three-phase workflow with real validation gates, copy-paste-ready prompts and commands, and a clean overview-to-references split with all bundle paths verified. The only nit is minor duplication between the Rules and Anti-Patterns sections and one repeated tagline, which costs a little token efficiency without harming clarity.

Suggestions

Merge the 'Rules' and 'Anti-Patterns To Reject' sections — both already forbid averaging the scores and softening the verdict — or move the anti-patterns into a single list to cut ~10 lines of overlap.

Drop the duplicate tagline: 'The panel does the depth; the Judge does the decision' appears both in the portability note at the top and again in Rules.

Dimension	Reasoning	Score
Conciseness	The body is lean — panelist mandates and commands earn their tokens and it never explains concepts Claude already knows — but has minor trimmables: 'The panel does the depth; the Judge does the decision' appears twice, and 'Anti-Patterns To Reject' partially restates 'Rules' (both forbid averaging and softening).	4 / 5
Actionability	Fully executable end to end: copy-paste bash commands whose flags match the scripts' actual argparse interfaces, five verbatim panelist prompts ready to paste with the [brief] slot, and an exact output template for the verdict — concrete guidance covering the common cases.	5 / 5
Workflow Clarity	A clearly sequenced three-step process with explicit checkpoints at each transition: brief_builder flags missing inputs before spending five subagents (a fix-then-proceed loop), a skip rule for 'just run it', 'Once all five return' as the join point, and a hard gate to run the synthesizer rather than average. Not the 4 anchor because validation steps are explicit, not implicit.	5 / 5
Progressive Disclosure	The body is a clear overview with well-signaled, one-level-deep references — three canons and two assets, each with a one-line description of what it contains, plus a Tooling table for the three scripts; all eight bundle paths exist and no reference nests further.	5 / 5
	Total	19 / 20 Passed

Description

96%Weight 40%Scale 1-5

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

An excellent description: third-person voice, explicit 'Use when' triggers with abundant natural synonyms including the slash command, and a concrete statement of the mechanism and output (5-role panel, GO/RESHAPE/KILL verdict, cheapest test). The only weakness is that a couple of broad trigger phrases leave slight routing ambiguity against sibling business-advisory skills.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete actions — 'Spins up a 5-angle panel (Critic, Champion, Analyst, Investigator, Customer)', 'a Judge returns one GO / RESHAPE / KILL verdict with the cheapest test to de-risk it' — covering what the skill does comprehensively, matching the top anchor rather than leaving minor gaps.	5 / 5
Completeness	Both halves are explicit: an explicit trigger clause ('Use when someone asks to roast an idea… or says "/roast"') and an explicit what ('Spins up a 5-angle panel… a Judge returns one GO / RESHAPE / KILL verdict').	5 / 5
Trigger Term Quality	It covers natural trigger phrases and their synonyms — 'roast an idea, pressure-test or stress-test an idea, validate a business idea, "convene the panel", get a brutal second opinion', plus the '/roast' slash-command form — comprehensive coverage of what users would actually say.	5 / 5
Distinctiveness Conflict Risk	Triggers like 'convene the panel' and '/roast' are highly distinct, but the broader phrases 'validate a business idea' and 'second opinion before building' carry minor overlap risk with closely related business-validation skills; disambiguation from siblings lives in the body's 'Distinct From' section rather than the description itself.	4 / 5
	Total	19 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: alirezarezvani/claude-skills
Path: productivity/roast/skills/roast/SKILL.md
Commit: 19392f7

Reviewed: 1 hour 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.