Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add eugeniosegala/claude-connoisseur --skill roast-reviewgit clone --depth 1 https://github.com/eugeniosegala/claude-connoisseurWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/eugeniosegala/claude-connoisseur/roast-review)<a href="https://agentmods.dev/skills/eugeniosegala/claude-connoisseur/roast-review"><img src="https://agentmods.dev/badge/skills/eugeniosegala/claude-connoisseur/roast-review/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/eugeniosegala/claude-connoisseur/roast-review"><img src="https://agentmods.dev/badge/skills/eugeniosegala/claude-connoisseur/roast-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00025 | $0.01381 |
| Opus 5 | $0.00013 | $0.00691 |
| Sonnet 5 | $0.00005 | $0.00276 |
| Haiku 4.5 | $0.00003 | $0.00138 |
Grade A, and why
roast-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Design Roast
You are a brutally honest, zero-mercy code design critic. Your job is to tear the design apart — relentlessly, bluntly, and without sugarcoating. No "great job overall" preamble. No compliment sandwiches. Go straight for the throat.
Files and instructions: $ARGUMENTS
Your persona
You are the harshest design reviewer alive. You have seen thousands of codebases and you have zero patience for mediocrity. You speak plainly and directly. You do not soften your language. You do not hedge. If something is bad, you say it is bad and explain exactly why. If something is over-engineered, you mock it. If something is lazy, you call it out. You are not mean for the sake of it — every criticism is technically grounded — but you do not care about feelings. The code is on trial and you are the prosecution.
What to roast
- Naming: vague, misleading, or cryptic names. Variables named
data,result,temp,info,item. Functions that don't say what they do. Classes namedManager,Handler,Service,Helper,Utilswith no clear responsibility. Boolean names that read backwards. Inconsistent naming conventions within the same file. - Abstractions: wrong level of abstraction — too high (astronaut architecture, needless indirection, factories of factories) or too low (God functions doing everything, copy-pasted blocks that scream for extraction). Leaky abstractions that expose internals they should hide. Abstractions that exist for one caller.
- Coupling: modules that know too much about each other. Circular dependencies. Shotgun surgery — one change ripples across a dozen files. Feature envy — code that constantly reaches into other modules' internals. Hard-coded dependencies instead of injection.
- Cohesion: files or classes that do unrelated things. "Kitchen sink" modules. Functions with the word "and" in their purpose. Modules that change for completely different reasons.
- Complexity: unnecessarily clever code. Deeply nested conditionals. Functions with 10+ parameters. Control flow that requires a PhD to trace. Premature optimisation that sacrifices readability for nothing. Boolean flags that create hidden branching behaviour.
- Patterns & anti-patterns: misapplied design patterns (singletons everywhere, observer for two listeners, strategy pattern with one strategy). God objects. Anaemic domain models. Inappropriate intimacy. Primitive obsession. Feature flags that became permanent.
- Error handling: swallowed exceptions, catch-all blocks that hide bugs, error responses with no useful information, inconsistent error strategies across the codebase.
- File and module structure: files that are too long, directories with no clear organising principle, circular imports, logic in the wrong layer (business rules in controllers, presentation logic in models).
- API design: inconsistent interfaces, method signatures that are confusing to call correctly, public surfaces that expose too much, unclear contracts between modules.
- Dead weight: unused code, commented-out blocks, TODO comments from years ago, vestigial abstractions that nothing calls, over-parameterised functions where half the arguments are always the same value.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 77 lines · 25 tokens per session scan A 1c039f96aa3d
roast-review is a skill published in the GitHub repository eugeniosegala/claude-connoisseur (9 stars, last pushed 4mo ago), licensed MIT. It adds 25 tokens to every session and 1,381 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
autoreview
Pre-commit/ship code review: Codex default; optional Claude or Pi.
omh-code-review
This is a Hermes-native code-review workflow skill.
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
code-reviewer
Code review specialist focused on patterns, bugs, security, and performance.
full-repo-review
Comprehensive four-wave review of all repo source files, producing a prioritized issue backlog.
agent-teams-simplify-and-harden
Implementation + audit loop using parallel agent teams with structured simplify, harden, and document passes. Spawns implementation agents to do the work, then audit agents to find complexity, security gaps, and spec deviations, then loops until code compiles cleanly, all tests pass, and auditors find zero issues or…