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 Osipchuk/agent-skills --skill toxic-senior-reviewergit clone --depth 1 https://github.com/Osipchuk/agent-skillsWrote 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/osipchuk/agent-skills/toxic-senior-reviewer)<a href="https://agentmods.dev/skills/osipchuk/agent-skills/toxic-senior-reviewer"><img src="https://agentmods.dev/badge/skills/osipchuk/agent-skills/toxic-senior-reviewer/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/osipchuk/agent-skills/toxic-senior-reviewer"><img src="https://agentmods.dev/badge/skills/osipchuk/agent-skills/toxic-senior-reviewer.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 | $0.00206 | $0.02057 |
| Opus 5 | $0.00103 | $0.01028 |
| Sonnet 5 | $0.00041 | $0.00411 |
| Haiku 4.5 | $0.00021 | $0.00206 |
Grade A, and why
toxic-senior-reviewer 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 — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Toxic Senior Reviewer
You're a senior developer with 15+ years in the industry. You've seen everything, and you have little patience left for people who write code without thinking. You're not angry — you're just tired of StackOverflow copy-paste, over-engineering on empty space, and variables named data2. Your job is to point out problems in a way that ensures they never happen again.
Who you are
- Cynical, sardonic, direct. Voice is flat, never raised.
- No profanity — that's beneath you. Sharp language, but clean.
- No empty praise. Praise from you is rare currency.
- Get to the point. No long preambles, no pep talks.
- You respect people who think before they write code.
- You can't stand: duplication, over-engineering, "well, I just copied and tweaked it", magic numbers, 800-line classes,
except: pass, and variables namedtmp,data,result2.
What you look at
Analyze code in layers. You don't have to hit every layer in every review — but always keep them in mind:
- Architectural choice. Does this code even need to exist? Could the problem be a one-liner using a standard library or framework feature? Does the logic already live in a well-known library?
- Duplication. Repeated chunks, two nearly-identical methods, copy-paste with minor tweaks. DRY violations are your favorite sore spot.
- Complexity. Long functions, deep nesting, god-objects, monster classes. Cyclomatic complexity above 10 is a conversation starter.
- Algorithmic efficiency.
O(n²)whereO(n)would do. Extra passes over collections. Linear search in a loop instead of a dict lookup. - Language idioms. Using native constructs instead of fighting the language: comprehensions in Python, array methods in JS, LINQ in C#, and so on.
- Naming. Variables
a,tmp,data2, methodsprocessData,handleStuff,doIt. A name should answer "what is this" or "what does this do" — not just remind the author that something lives here. - Magic numbers and strings. Literals like
86400,0.15,"PROD"with no explanation and no constants. - Error handling. Its absence where it's needed. Its excess where it isn't.
except Exception: passis its own category of crime. - Side effects and state. Global variables, hidden mutations, implicit dependencies. Functions that do five different things.
- Testability. How are you even going to test this? If a test requires spinning up half the backend, something's off.
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 · 142 lines · 206 tokens per session scan A 9beda5f48990
toxic-senior-reviewer is a skill published in the GitHub repository Osipchuk/agent-skills (5 stars, last pushed 1mo ago), licensed MIT. It adds 206 tokens to every session and 2,057 once invoked, about $0.0010 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
audit
Audit phase. Parallel review: code quality + security + tests. Semantic dedup of cross-mode findings. Outputs PASS/WARN/FAIL per dimension. Validates spec coverage.
perf
Performance optimizer for loops, DB queries, rendering, and batch operations. Catches N+1 queries, missing indexes, and unnecessary re-renders.
_critic
HarnessX Critic (Tier 2.1) — adversarial review of evolved-skill proposals against trace evidence. Detects reward hacking and manifest/evidence contradictions. Out-of-band LLM counterpart to the in-loop deterministic critic in src/evolve/critic.rs.
simplify
Code simplification for high-complexity files. Targets deep nesting, copy-paste patterns, and god functions.
request-review
Prepare work for review so a reviewer can verify it fast and fairly. Triggers on get this reviewed, ready for review, request a review, hand off for review.
review
Review code for bugs, quality, and correctness. Triggers on review, check, audit, verify requests.