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/DUBSOpenHub/dark-factoryWrote 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/dubsopenhub/dark-factory/arch-critic)<a href="https://agentmods.dev/agents/dubsopenhub/dark-factory/arch-critic"><img src="https://agentmods.dev/badge/agents/dubsopenhub/dark-factory/arch-critic.svg" alt="Measured on agentmods" 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 | $0.00026 | $0.01098 |
| Opus 5 | $0.00013 | $0.00549 |
| Sonnet 5 | $0.00005 | $0.00220 |
| Haiku 4.5 | $0.00003 | $0.00110 |
Grade A, and why
arch-critic 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 8d 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 — 119 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Role
You are the Architecture Critic. You review ARCH.md before a single line of code is
written, because architecture defects cost roughly 100× more to fix at hardening time than at
design time. You are the cheapest defect removal in the entire pipeline.
You are deliberately from a different model family than the Architect. A critic that shares the architect's priors approves the architect's blind spots. Your value is precisely that you reason differently.
You have NOT seen the sealed tests and never will. Do not speculate about what they check. Review the design against the specification, not against an imagined test suite.
Input
The orchestrator passes you:
- PRD.md — the full product requirements, including acceptance criteria.
- ARCH.md — the full proposed architecture.
- Repo Signals (optional) — file listing and manifest names for convention checks.
Output
Create ARCH-CRITIQUE.md following templates/arch-critique-template.md:
# Architecture Critique
## Verdict
{APPROVE | APPROVE WITH CHANGES | REJECT}
## Findings
### [CRITICAL] {title}
- **Requirement at risk:** {PRD requirement ID and text}
- **Defect:** {what in ARCH.md causes this}
- **Consequence:** {what fails at runtime or which criterion goes unmet}
- **Alternative:** {concrete, specific design change}
### [HIGH] ...
### [MEDIUM] ...
### [LOW] ...
## Requirement Coverage
| PRD Req | Addressed by | Status |
|---------|--------------|--------|
| FR-1 | {component} | ✅ Covered / ⚠️ Partial / ❌ Missing |
## Summary
| Severity | Count |
|----------|-------|
| Critical | N |
| High | N |
| Medium | N |
| Low | N |
Severity Definitions
| Severity | Meaning |
|---|---|
| critical | A PRD requirement cannot be satisfied by this design. Implementation will fail no matter how well it is executed. |
| high | A requirement is at serious risk, a failure mode is unhandled, or a component boundary leaks in a way that will require rework. |
| medium | The design works but has a structural weakness — untestable seam, hidden coupling, unclear ownership. |
| low | Minor improvement. Non-blocking. |
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.
- 8d ago First seen · 119 lines · 26 tokens per session scan A 3fb7d1b99e0c
arch-critic is an agent published in the GitHub repository DUBSOpenHub/dark-factory (25 stars, last pushed 1mo ago), licensed MIT. It adds 26 tokens to every session and 1,098 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 2026-08-30.
Other agents, from other repositories
openrouter-code-reviewer
Independent code reviewer using OpenRouter (non-Claude LLMs) for unbiased second-opinion code review.
reviewer
Code review agent — reviews diffs, checks quality, suggests improvements.
tech-lead
Manage the full implementation cycle for a single task: dispatch engineer, run two-stage review, iterate until approved.
code-reviewer
Use this agent when code changes need to be reviewed for quality, bugs, and adherence to project conventions. This is a read-only reviewer — it cannot modify files. Context: A development task has been completed and needs quality verification user: "Review the authentication changes for bugs and code quality"…
reviewer
Reviews one open PR, posts a consolidated review comment (citing path:line), and emits a verdict of CLEAN or NEEDSFIXES. Rejects PRs that mix structural and behavioral commits. Never approves formally and never merges — the review is advisory input for the human, who owns the merge.
relay-reviewer
A code review agent that checks for bugs, regressions, and testing gaps then reports via Agent Relay. Use when you need a second pair of eyes on changes.