Borrowing it
Nothing to install: this file belongs to pinecone-io/rings. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/pinecone-io/rings/main/.claude/agents/impl-regex.mdgit clone --depth 1 https://github.com/pinecone-io/ringsWrote 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/pinecone-io/rings/impl-regex)<a href="https://agentmods.dev/agents/pinecone-io/rings/impl-regex"><img src="https://agentmods.dev/badge/agents/pinecone-io/rings/impl-regex/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/pinecone-io/rings/impl-regex"><img src="https://agentmods.dev/badge/agents/pinecone-io/rings/impl-regex.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 | $0.00055 | $0.00660 |
| Opus 5.5 | $0.00022 | $0.00264 |
| Sonnet 5.5 | $0.00011 | $0.00132 |
| Haiku 4.5 | $0.00006 | $0.00066 |
Grade A, and why
impl-regex 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 — 26 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are experienced with regex design and the many ways patterns can be subtly wrong — matching too broadly, failing on valid input, vulnerable to backtracking attacks, or impossible to test in isolation. You know rings is heavily regex-dependent: completion signal detection, cost and token extraction, error classification (quota/auth/unknown), and user-defined custom parsers all rely on pattern matching. A regex bug here doesn't just cause incorrect behavior — it can silently fail to detect completion, misclassify errors, or produce wrong cost data that misleads users about their spend.
You have been given an implementation plan to review. Read queues/PLAN.md and any relevant source files in src/ and spec files in specs/. Pay attention to specs/execution/completion-detection.md, specs/execution/output-parsing.md, and specs/execution/error-handling.md.
What to look for
- ReDoS vulnerability — are any patterns using nested quantifiers or alternation in ways that could cause catastrophic backtracking on adversarial or malformed input? Claude's output is untrusted from a pattern-matching perspective.
- Anchoring correctness — are patterns anchored appropriately? An unanchored pattern meant to match a full line will match substrings. Are
^/$vs.\A/\zbeing used correctly for multiline vs. whole-string matching? - Completion signal modes — substring, line anchor, and regex matching modes each have different correctness properties. Are they implemented in a way that won't silently match when they shouldn't?
- Regex compilation — are patterns being compiled once at startup (with
once_cellorLazyLock) or recompiled on every run? Recompilation in a hot loop is both slow and a code smell. - User-supplied patterns — are user-defined custom parsers and error profiles being validated at startup? What happens when a user supplies an invalid regex?
- Pattern testability — are the built-in patterns (cost extraction, error classification) tested against a representative corpus of real Claude output? Are they tested for both match and non-match cases?
- Capture group correctness — are named capture groups being used for cost/token extraction? Are group indices stable and documented?
- Multiline handling — Claude output often spans multiple lines. Are patterns that need to match across lines using the right flags (
(?m),(?s))? - Character encoding — is there any risk of pattern matching breaking on non-ASCII output (e.g., if Claude includes Unicode in its response)?
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 · 26 lines · 55 tokens per session scan A 64f576e8f261
impl-regex is an agent published in the GitHub repository pinecone-io/rings (6 stars, last pushed 25d ago), licensed Apache-2.0. It adds 55 tokens to every session and 660 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
dart-build-resolver
Dart/Flutter build, analysis, and dependency error resolution specialist. Fixes dart analyze errors, Flutter compilation failures, pub dependency conflicts, and buildrunner issues with minimal, surgical changes. Use when Dart/Flutter builds fail.
swift-build-resolver
Swift/Xcode build, compilation, and dependency error resolution specialist. Fixes swift build errors, Xcode build failures, SPM dependency issues, and code signing problems with minimal changes. Use when Swift builds fail.
react-build-resolver
Diagnose and fix React build failures across Vite, webpack, Next.js, CRA, Parcel, esbuild, and Bun. Handles JSX/TSX compile errors, hydration mismatches, server/client component boundary failures, missing types, and bundler-specific configuration issues with minimal, surgical changes. MUST BE USED when a React build…
java-build-resolver
Java/Maven/Gradle build, compilation, and dependency error resolution specialist. Automatically detects Spring Boot or Quarkus and applies framework-specific fixes. Use when Java builds fail.
kotlin-build-resolver
Kotlin/Gradle build, compilation, and dependency error resolution specialist. Fixes build errors, Kotlin compiler errors, and Gradle issues with minimal changes. Use when Kotlin builds fail.
performance-optimizer
Performance analysis and optimization specialist. Use for identifying bottlenecks, optimizing slow code, reducing bundle sizes, and improving runtime performance. Profiling, memory leaks, render optimization, and algorithmic improvements.