Redis' own guidance for FT.CREATE schema design, FT.SEARCH / FT.AGGREGATE / FT.HYBRID, HNSW vector similarity and RAG retrieval pipelines.
Claude Code Settings Audit
Summary
Analyse a repository's tech stack and generate a recommended Claude Code settings.json permission allowlist for its read-only commands.
Features
- Detects the stack from lockfiles and build manifests, not guesswork
- Covers Python, Node, Go, Rust, Ruby, Java, Docker and Terraform
- Recognises monorepo layouts and adjusts recommendations
- Restricts the generated allowlist to read-only commands
- Works as a first-time setup step or an audit of existing settings
Install This Skill
Add this skill to your favorite AI agent in a few steps.
Skill Content
Usage Instructions
Learn how to use this skill with different AI agents.
Example Usage
Audit this repository and recommend Claude Code settings.json permissions.
Description
Every new project starts with the same friction: the agent asks permission for pnpm check, then cargo test, then terraform validate, and you approve each one by hand. This skill from Sentry's collection does the survey for you and produces a settings.json permissions block tailored to what the repository actually contains.
It begins by detecting the stack from indicator files rather than guessing — pyproject.toml, uv.lock and requirements.txt for Python; package.json and the various lockfiles for Node; go.mod, Cargo.toml, Gemfile, pom.xml and Gradle build files; Makefile, Dockerfile and docker-compose.yml for build tooling; *.tf, kubernetes/ and helm/ for infrastructure. It also recognises monorepo markers (turbo.json, nx.json, lerna.json, pnpm-workspace.yaml), which matter because they change what the safe command surface looks like.
The output is scoped to read-only commands. That is the point: allowlisting git status, pnpm test or terraform plan removes almost all of the prompt noise without handing an agent the ability to mutate anything, so the permission dialogs that remain are the ones actually worth reading.
Useful when onboarding a repository to Claude Code, and equally useful as an audit of settings that accumulated by hand over time and now allow more than anyone intended. Published by Sentry in getsentry/sentry-skills under the open Agent Skills format.
Related Skills
Supabase's official skill covering Database, Auth, Edge Functions, Realtime, Storage, Vectors, Cron and Queues — with a hard rule to verify against the live changelog before writing code.
GreenSock's official ScrollTrigger skill — scroll-linked animation, pinning, scrub and trigger positioning, with the exact start/end syntax agents get wrong.
Microsoft's official skill that turns an agent into an Azure solution architect — 10 design principles, 6 architecture styles, 44 design patterns and a Well-Architected review workflow.