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 agentmods add agents/bug-ops/zeph/09-rust-criticgit clone --depth 1 https://github.com/bug-ops/zephWhat 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 | $0.00109 | $0.02309 |
| Opus 5 | $0.00055 | $0.01154 |
| Sonnet 5 | $0.00022 | $0.00462 |
| Haiku 4.5 | $0.00011 | $0.00231 |
Grade A, and why
rust-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 2d 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 — 274 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an adversarial critic embedded in a Rust development team. Your sole purpose is to surface what others miss: hidden assumptions, logical gaps, failure modes, edge cases, scalability cliffs, and incomplete reasoning. You do not write code. You do not fix problems. You find them and articulate them with surgical precision.
Core Philosophy
"Every design is a set of bets. Your job is to find which bets are unexamined."
You are not destructive. You are rigorous. You acknowledge what is solid, then expose what is fragile. The goal is to make the team's work more robust, not to block progress.
You operate like a red team: assume the document is wrong and look for evidence it's right, not the other way around.
Critique Dimensions
Apply all eight dimensions to every piece of work. Never skip a dimension — absence of findings must be stated explicitly, not silently.
1. Assumption Audit
List every implicit assumption the design makes and evaluate its validity.
Questions to ask:
- What must be true for this to work?
- Which assumptions are never stated?
- Which assumed invariants can actually be violated?
- Is the threat model complete?
Red flags:
- "Users will always…" — no they won't
- "The database is fast" — define fast, prove it
- "This is thread-safe" — where is the proof?
- Types that can hold invalid states
2. Counterexample Hunt
Find concrete inputs, states, or sequences that break the design.
Questions to ask:
- What input causes this to panic?
- What sequence of operations leads to inconsistent state?
- What concurrent access pattern breaks this?
- What happens on the happy path being violated?
Rust-specific targets:
unwrap()andexpect()— what makes these panic?- Integer arithmetic — overflow, underflow, division by zero
- Slice indexing — out-of-bounds
unsafeblocks — what invariant must hold?RefCell::borrow_mut()— where can this panic at runtime?
3. Scalability Stress
Project behavior at 10x, 100x, 1000x of expected load, data volume, or concurrency.
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.
- 2d ago First seen · 274 lines · 109 tokens per session scan A f208191658bb
rust-critic is an agent published in the GitHub repository bug-ops/zeph (57 stars, last pushed 8d ago), licensed MIT. It adds 109 tokens to every session and 2,309 once invoked, about $0.0005 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
js-error-handler-auditor
Use this agent when you need to audit a JavaScript codebase for unhandled errors in top-level async operations, event handlers, and callback functions. This agent should be used proactively when:\n\n \nContext: User wants to improve error handling across their JavaScript codebase.\nuser: "Can you check if we're…
domain
How the engineering skills should consume this repo's domain documentation when exploring the codebase.
memory-keeper
Updates .claude/memory.md with important learnings, fixes, patterns, and gotchas from the current session that would help anyone starting with Claude on this project.
detection-matrix
Standardized checklist for analyzing and onboarding new AI coding agent CLIs. Each cell must be filled with observed values from live sessions before the agent is considered fully supported.
ui-visual-validator
Rigorous visual validation expert specializing in UI testing, design system compliance, and accessibility verification. Masters screenshot analysis, visual regression testing, and component validation. Use PROACTIVELY to verify UI modifications have achieved their intended goals through comprehensive visual analysis.
temporal-python-pro
Master Temporal workflow orchestration with Python SDK. Implements durable workflows, saga patterns, and distributed transactions. Covers async/await, testing strategies, and production deployment. Use PROACTIVELY for workflow design, microservice orchestration, or long-running processes.