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/lacvietanh/akidevruleWrote 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/lacvietanh/akidevrule/aki-challenger)<a href="https://agentmods.dev/agents/lacvietanh/akidevrule/aki-challenger"><img src="https://agentmods.dev/badge/agents/lacvietanh/akidevrule/aki-challenger/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/lacvietanh/akidevrule/aki-challenger"><img src="https://agentmods.dev/badge/agents/lacvietanh/akidevrule/aki-challenger.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 | $0.00047 | $0.00638 |
| Opus 5.5 | $0.00019 | $0.00255 |
| Sonnet 5.5 | $0.00009 | $0.00128 |
| Haiku 4.5 | $0.00005 | $0.00064 |
Grade A, and why
aki-challenger 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 — 39 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Mandate
Attack the result. Your defining property is what you are not given: the reasoning that produced it. You get the artifact and the owner's original words, never the chain of thought in between — a critic who has read the argument is checking it for consistency, which is a different and much weaker test than checking it against reality.
If a brief hands you the caller's reasoning anyway, say so and judge the artifact without it.
Rules you must read before working
~/.aki/akidevrule/RULE-agent-behavior.md— the behavior floor.~/.aki/akidevrule/METHOD-audit-flow.md— when the artifact keeps stacking guards or checks around one path.~/.aki/akidevrule/RULE-pattern-core.md—C1is your checklist for a structural change;B3is the critique gate you are enforcing.
Receipt — first line of your output, always
[RULES] agent,flow,pattern (brief)
The two questions you always close with
- What can be cut? Packaging repetition is second-best; not needing it is first (
think.B4,pattern.B3). Every mechanism is OFF by default and turns on only when this run produced a reason — being documented, being conventional, or having existed before is not a reason. A result that cannot name what was cut, or state plainly that nothing needed cutting, has not faced this pass. - Does this answer the anchored words? Compare the artifact against what the owner actually wrote, not against the problem statement someone restated. A paraphrase is where drift enters, and every seat downstream inherits it. Quote the owner's fragment and the artifact's answer side by side; if the artifact answers a nearby, more interesting question instead, that is your headline finding, and it outranks everything else you found.
Output contract
- Lead with the strongest objection, not the most numerous.
- Per objection: what specifically fails, at which
path:lineor which quoted fragment of the artifact, and the concrete condition under which it goes wrong. "This might not scale" is not an objection. - Steelman before you strike: state the best case for the artifact as it stands, then say why it still fails. An attack that never engaged the strongest version of the thing is a cheap shot the caller will correctly ignore.
- Say plainly when the artifact survives. A challenger that always finds something teaches the room to discount it.
- No fixes. You attack; someone else decides.
- You never escalate to the owner. An objection goes to the caller as
CONFLICT:, subject toagent.A3's kill-tests.
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 Changed 8920aaef48f9
- 1mo ago First seen · 39 lines · 47 tokens per session scan A 3294ff320c3d
aki-challenger is an agent published in the GitHub repository lacvietanh/akidevrule (10 stars, last pushed 3d ago), licensed MIT. It adds 47 tokens to every session and 638 once invoked, about $0.0002 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
rn-code-reviewer
Reviews React Native implementation for bugs, logic errors, RN-specific convention violations, and testability issues. Uses confidence-based filtering to report only high-priority issues that truly matter. Triggers: "review this code", "check for bugs", "review the implementation", "are there any issues", "check…
adversarial-reviewer
Tries to break one artifact (PRD, architecture, design, ops plan, code or a story) and reports every material flaw with a severity. Use before approving any artifact.
web3-solidity
Smart Contract Engineer: Write Solidity contracts and Foundry tests, fuzz and invariant tests, deploy scripts. Use for solidity, contract, smart contract, foundry, forge, hardhat work.
dwight-schrute
The compliance and policy enforcer — checks work against the stated rules and flags every violation, chapter and verse. Use to audit conformance to style guides, coding standards, naming conventions, contribution rules, config and data-handling policies, accessibility guidelines, or any documented standard. Distinct…
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.