performance agents

65 tagged performance, measured the same way as everything else here.

Browse within: claude-plugin 28agent-team 24ai-assisted-development 12coding 12vibe 12vibecoding 12typescript 10Optimization 9lightweight 8virtual-list 8virtual-scroll 8virtualization 8anthropic-claude 7marketplace 5

code-review

02

floor/vlist

Agent Claude Code

Review code against vlist project rules and conventions. Use proactively after writing or editing code to catch violations before committing.

16 1mo ago A 26 tokens MIT

dev

03

floor/vlist

Agent Claude Code

Implement features, fix bugs, and iterate until all quality checks pass. Use for autonomous development tasks that benefit from worktree isolation and self-correction.

16 1mo ago A 32 tokens MIT

perf-audit

04

floor/vlist

Agent Claude Code

Audit vlist source code for hot-path performance violations. Use when changes touch scroll handling, rendering, range calculations, or any code that runs per-frame.

16 1mo ago A 35 tokens MIT

deep-bug-scanner

05

bradwindy/ultimate-code-review

Agent

Use this agent for deep bug detection that traces full call graphs of changed functions. Unlike shallow scanning, this agent follows execution paths through all layers to find bugs that only manifest when the full call chain is considered. Context: A PR changes a validation function used by multiple API endpoints.…

2 4mo ago A 198 tokens original MIT

devils-advocate

06

bradwindy/ultimate-code-review

Agent

Use this agent to adversarially challenge every finding in the synthesized report. Verifies technical claims against the web, checks for false positives, and assigns confidence assessments. Context: The synthesizer has produced a unified report with 15 findings. user: "Challenge these findings" assistant: "I'll use…

2 4mo ago A 109 tokens original MIT

bradwindy/ultimate-code-review

Agent

Use this agent to detect out-of-scope code changes in a PR/MR. Reads the PR/MR description and any linked tickets/issues, then assesses whether each changed file and hunk is related to the stated purpose. Cautious by default -- only flags changes when quite confident they are unrelated. Context: A PR titled "Fix login…

2 4mo ago A 235 tokens original MIT