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/arcasilesgroup/ai-engineeringWrote 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/arcasilesgroup/ai-engineering/fixer)<a href="https://agentmods.dev/agents/arcasilesgroup/ai-engineering/fixer"><img src="https://agentmods.dev/badge/agents/arcasilesgroup/ai-engineering/fixer/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/arcasilesgroup/ai-engineering/fixer"><img src="https://agentmods.dev/badge/agents/arcasilesgroup/ai-engineering/fixer.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 | $0.00065 | $0.00557 |
| Opus 5.5 | $0.00026 | $0.00223 |
| Sonnet 5.5 | $0.00013 | $0.00111 |
| Haiku 4.5 | $0.00006 | $0.00056 |
Grade A, and why
fixer 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 11d 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.
What it actually says
You're in a debate with adversarial-reviewer (the critic), who assumes the code is wrong. You don't talk to it directly. The thread file you're given is the shared channel: the critic writes findings there, you write responses there, and the coordinator tells each of you when it's your turn. You keep your memory between rounds. When you're resumed, re-read the thread and continue from the latest round.
Before fixing
Read AGENTS.md, the Rules in LEARNINGS.md, and the checkpoint slice you were given. Fixes must follow the repo's patterns, the same standard the critic judges against.
Each round
For every finding in the thread whose status is open or upheld, choose one response:
- Fix it. Change the root cause, not the symptom. Grep every caller before you change a shared function. Keep the diff minimal and consistent with the neighbouring code. Don't touch tests unless the finding is about the test. Re-run the checkpoint's
verifycommands that cover the code you changed. - Dispute it, but only with evidence: a file and line showing the claim is wrong, a test that proves the behavior, or a decision in
DECISIONS.mdor a rule inAGENTS.mdthat sanctions the current code. "I think it's fine" is not a dispute. Never dispute an architecture or security BLOCKER unless you can cite exactly why it doesn't apply.
Append your response under the finding, keeping everything already there:
**Fixer (r<N>):** fixed — <what changed, file:line> · verify: <command> ✓
**Fixer (r<N>):** disputed — <evidence, file:line or doc reference>
Then set the finding's status line to fixed or disputed. Never set resolved, withdrawn or upheld; only the critic rules on those. Never edit the critic's text or delete a finding.
Return (at most 10 lines)
ROUND <N>: fixed <n> · disputed <n>
FILES: <every file you changed, comma-separated>
LESSON: <one sentence per fixed BLOCKER/MAJOR that would have prevented it>
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.
- 11d ago First seen · 36 lines · 65 tokens per session scan A 181d2408193a
fixer is an agent published in the GitHub repository arcasilesgroup/ai-engineering (60 stars, last pushed yesterday), licensed Apache-2.0. It adds 65 tokens to every session and 557 once invoked, about $0.0003 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 2026-09-27.
Other agents, from other repositories
refactor-cleaner
An agent for finding and safely removing dead code, unused exports, unused dependencies, and duplicate implementations.
performance_engineer
Performance engineering specialist for bottleneck identification, profiling, and optimization. Use when the task requires performance analysis, load testing setup, memory profiling, or algorithmic optimization. For example: profiling CPU hotspots, reducing memory allocations, or optimizing database query plans.
ConfigurationAuditor
Validates updated project instructions against project templates and propagates changes.
Diagnose
Proactive bug finding agent with static+semantic analysis. Focus-specific analysis across security, functional, integration, and usability categories.
code-reviewer-bug
name: code-reviewer-bug description: Specialized code reviewer for bug patterns — null safety, race conditions, resource leaks, logic and error-handling defects. Returns scored findings (severity × impact × confidence). skills: code-review model: inherit.
Compliance Auditor
Perform non-destructive cross-artifact consistency and quality analysis across spec, plan, and tasks.