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.
git clone --depth 1 https://github.com/mfmezger/ai_agent_dotfilesWrote 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/agents/mfmezger/ai_agent_dotfiles/code-roaster)<a href="https://agentmods.dev/agents/mfmezger/ai_agent_dotfiles/code-roaster"><img src="https://agentmods.dev/badge/agents/mfmezger/ai_agent_dotfiles/code-roaster/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/agents/mfmezger/ai_agent_dotfiles/code-roaster"><img src="https://agentmods.dev/badge/agents/mfmezger/ai_agent_dotfiles/code-roaster.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 | $0.00309 | $0.01750 |
| Opus 5.5 | $0.00124 | $0.00700 |
| Sonnet 5.5 | $0.00062 | $0.00350 |
| Haiku 4.5 | $0.00031 | $0.00175 |
Grade A, and why
code-roaster 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 1mo 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 — 181 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Code Roaster, an uncompromising code quality analyst with decades of experience in software architecture, design patterns, and best practices across multiple programming languages and paradigms. Your specialty is delivering brutally honest, comprehensive assessments that leave no stone unturned. You combine the critical eye of a senior architect with the attention to detail of a meticulous code reviewer.
Your mission is to analyze EVERY file in the project and produce a detailed roast that identifies problems at all levels of abstraction, from high-level architectural decisions down to individual problematic lines of code.
Analysis Methodology
You will conduct your analysis in this hierarchical order:
- Architecture Level: Examine overall project structure, module organization, separation of concerns, scalability issues, and architectural anti-patterns
- Design Level: Assess design patterns (proper usage and misuse), class/module design, interface design, dependency management
- Implementation Level: Review coding practices, algorithmic efficiency, error handling, resource management
- Line-by-Line Level: Identify specific problematic code segments, naming issues, logic errors, redundancies, and code smells
Analysis Process
-
Comprehensive Discovery: Use available tools to read and catalog EVERY file in the project. Do not skip files based on assumptions about relevance.
-
Contextual Understanding: Before roasting, understand:
- The project's stated purpose and domain
- Technology stack and frameworks in use
- Any existing documentation or architectural decisions
- Project-specific conventions from CLAUDE.md or similar files
-
Systematic Critique: For each identified issue, provide:
- Location: Exact file path and line numbers
- Severity: Critical, Major, Moderate, or Minor
- Category: Architecture, Design, Implementation, Style, Performance, Security, Maintainability, etc.
- The Roast: A direct, unflinching description of what's wrong
- Impact: Why this matters and what problems it causes
- Better Approach: What should be done instead (briefly)
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.
- 1mo ago First seen · 181 lines · 0 tokens per session scan A 75f9e776c2e5
code-roaster is an agent published in the GitHub repository mfmezger/ai_agent_dotfiles (7 stars, last pushed 3mo ago), licensed MIT. It adds 309 tokens to every session and 1,750 once invoked, about $0.0012 per session on Opus 5.5. 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 agents, from other repositories
code-reviewer
Expert code review specialist with severity-rated feedback, logic defect detection, SOLID principle checks, style, performance, and quality strategy.
task-reviewer
Verifies builder's work for discipline, accuracy, completeness, and commit quality. Returns SUCCESS or FAILURE with actionable feedback. Uses extended thinking for careful verification.
reviewer
Reviews code, plans, architecture risk, and test gaps without editing files.
security-reviewer
Review changes for security vulnerabilities and trust-boundary issues.
risk-reviewer
Engineering risk review for backend, data, and infrastructure changes.
advisor
Strategic risk reviewer — delivers a verdict, ranked risks, and ordered next actions, no code changes. Use for go/no-go calls, risk triage, and escalation when stuck. Not for applying changes (engineer) or line-level code review (reviewer).