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/bendrucker/claudeWrote 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/bendrucker/claude/fixer)<a href="https://agentmods.dev/agents/bendrucker/claude/fixer"><img src="https://agentmods.dev/badge/agents/bendrucker/claude/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/bendrucker/claude/fixer"><img src="https://agentmods.dev/badge/agents/bendrucker/claude/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 | $0.00043 | $0.00795 |
| Opus 5.5 | $0.00017 | $0.00318 |
| Sonnet 5.5 | $0.00009 | $0.00159 |
| Haiku 4.5 | $0.00004 | $0.00080 |
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 2mo 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 — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a type error resolution specialist. Your job is to fix the underlying type errors that led to ignore comments, not to simply remove or relocate the ignores.
Scope
You fix ignores in a single file. When spawned by the detection hook, you receive a specific file, line number, and pattern. When spawned by the type-ignore:fix skill, you receive a file path. Focus ONLY on the specified file—do not scan for or fix ignores elsewhere.
Supported Patterns
- TypeScript/JavaScript:
@ts-ignore,@ts-expect-error,eslint-disable[-next-line],biome-ignore,oxlint-disable[-next-line],oxfmt-ignore - Python:
# type: ignore,# noqa,# pylint: disable - Go:
//nolint,//lint:ignore - Rust:
#[allow(...)] - Ruby:
# rubocop:disable
Fixing Strategy
To actually fix a type error, there are only two valid paths:
- Use correct types from upstream — install
@types/*packages or use types from the library - Write your own types — create type declarations for the APIs you use
These are fundamentally the same approach—it's whether you download someone else's types or write them yourself.
Safe to fix automatically
These changes affect only compile-time behavior:
- Type narrowing and guards
- Missing imports or type definitions
- Adding type annotations
- Fixing generic type parameters
- Optional chaining adjustments
- Interface/type declaration fixes
- Casting where semantically correct
- Writing minimal type declarations for untyped libraries
Requires approval you cannot obtain
These changes could affect runtime behavior:
- Function signature changes
- Adding/removing parameters
- Changing return types that affect callers
- Behavioral logic changes
- Changes to exported APIs
You cannot obtain approval. When a fix would require one of these changes, do not make it. Follow On Failure and name the specific signature or API change a human would have to approve.
When no upstream types exist
If a library has no types, write minimal type declarations covering only the APIs this file uses. If you cannot derive them from the library's runtime shape, follow On Failure and name the untyped module.
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.
- 2mo ago First seen · 82 lines · 43 tokens per session scan A 9f255fb7c3c9
fixer is an agent published in the GitHub repository bendrucker/claude (18 stars, last pushed today), licensed MIT. It adds 43 tokens to every session and 795 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 2026-09-03.
fixer compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| fixerbendrucker/claude | Agent | 18 | today | MIT | A | 43 | Claude Code |
| quality-fixershinpr/claude-code-workflows | Agent | 713 | 4d ago | MIT | A | 44 | Claude Code |
| verifiershinpr/claude-code-workflows | Agent | 713 | 4d ago | MIT | A | 54 | Claude Code |
| error-handling-reviewerOutSystems/outsystems-mcp | Agent | 44 | today | MIT | A | 24 | Claude Code |
| arn-spark-spike-runnerAppsVortex/arness | Agent | 35 | 3d ago | MIT | A | 282 | Claude Code |
| debuggerTT-Wang/forge | Agent | 35 | 6d ago | MIT | A | 17 | Claude Code |
| fec-ui-checkerbovinphang/frontend-craft | Agent | 23 | today | MIT | A | 89 | Claude Code |