FlashInfer is a library and kernel generator that supplies GPU operations used to run large language model inference, including attention, matrix multiplication, and mixture-of-experts computations. It helps engineers build and optimize LLM serving systems across supported GPU hardware and backend implementations. Its catalogue add-ons provide skills and instructions for working with FlashInfer.
Borrowing it
Nothing to install: this file belongs to flashinfer-ai/flashinfer. 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/flashinfer-ai/flashinfer/main/.claude/skills/self-review/SKILL.mdgit clone --depth 1 https://github.com/flashinfer-ai/flashinferWrote 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/skills/flashinfer-ai/flashinfer/self-review)<a href="https://agentmods.dev/skills/flashinfer-ai/flashinfer/self-review"><img src="https://agentmods.dev/badge/skills/flashinfer-ai/flashinfer/self-review/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/skills/flashinfer-ai/flashinfer/self-review"><img src="https://agentmods.dev/badge/skills/flashinfer-ai/flashinfer/self-review.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 | $0.00020 | $0.01000 |
| Opus 5.5 | $0.00008 | $0.00400 |
| Sonnet 5.5 | $0.00004 | $0.00200 |
| Haiku 4.5 | $0.00002 | $0.00100 |
Grade A, and why
self-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Self-Review Before Opening a PR
Run this when a change is being prepared for a PR, after it is complete and before writing the PR description. It walks the diff through the repository's published review guidance and PR rules, so that common issues are caught before a human looks and review time goes to the design. It is informal and optional: it exists to make review easier for reviewers and contributors, not to gate anything, and changes made only to test something locally do not need it. This skill does not restate the rules; it points at where they live and says how to apply them to a diff.
Step 1: Get the diff
git fetch upstream main 2>/dev/null || git fetch origin main
BASE=$(git merge-base HEAD upstream/main 2>/dev/null || git merge-base HEAD origin/main)
git status --short # untracked files that belong to the change must be added
git diff --stat "$BASE" # merge-base vs working tree: committed and uncommitted changes
git diff "$BASE"
Review the whole diff, not only the files you remember editing. Unrelated changes (submodule pointer drift, scratch files) must be left out.
Step 2: Apply the review guidance
Read docs/code_review_guidance.md and walk its
focus areas and checklist over the diff. In particular:
-
Kernel logic is in scope. Read indexing/stride math, boundaries and predication, accumulation dtype and scaling, barrier placement, alignment assumptions. Do not rely on a green test run as a substitute for reading the code.
-
Interfaces get copied. Check argument order, plan/run split, decorator use (
@flashinfer_api,@backend_requirement), naming, and thatinclude/stays Torch-free. -
Tests. New behavior and edge cases are covered, numerics have a reference check, and architecture guards are correct. Run the touched test files, not the whole suite:
pre-commit run --files $(git diff --name-only --diff-filter=d "$BASE") # skip deleted files pytest tests/<touched files>
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 First seen · 83 lines · 20 tokens per session scan A 0a7597177552
self-review is a skill published in the GitHub repository flashinfer-ai/flashinfer (6,763 stars, last pushed today), licensed Apache-2.0. It adds 20 tokens to every session and 1,000 once invoked, about $0.0001 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 2026-09-24.
Other skills, from other repositories
review-implement-phase
Implements triaged review actions, commits focused fixes, and posts Done plus resolves threads. Use when the user wants only the implementation phase of the review-framework workflow.
github-pr-workflow
Prepare a GitHub pull request from a feature branch — branch hygiene, commit shape, title/body, verification notes, screenshots for UI work, and replies to review comments.
review-delta
Review only changes since last commit using impact analysis. Token-efficient delta review with automatic blast-radius detection.
github-contributor
A step-by-step guide for contributing code, tests, documentation, or build changes to open-source projects you do not maintain. It explains how to prepare a pull request that follows the project’s rules and is easier for maintainers to review.
hero-diff-highlight
A dark code editor reveals a red deletion then a green addition with a soft pulse. Function: PR / code-change explainer.
review-trtmc-pr
Review a TensorRT-Model-Connect GitHub PR or local contributor branch against the current post-#1093 model-family isolation architecture, repository rules, behavioral correctness, third-party dependency and license obligations, and exact-head validation evidence. Use for contributor self-review before marking a PR…