valkey-iovalkey-io/valkey-glide/CLAUDE.md

valkey-glide

CLAUDE.md800 starsChanged 9 months agoApache-2.0View on GitHub

Valkey GLIDE is an official open-source client library for Valkey and Redis OSS. It uses a core driver written in Rust with language-specific wrappers for Python, Java, Node.js, and Go, and other languages in seperate repositories. Socket IPC wrappers → socket_listener → glide-core → Valkey/Redis FFI wrappers → glide-core → Valkey/Redis

In this repository8 other files it gives its agents.
- AGENTS.mdAGENTS.md
- glide-core/AGENTS.mdAGENTS.md
- go/AGENTS.mdAGENTS.md
- java/AGENTS.mdAGENTS.md
- node/AGENTS.mdAGENTS.md
- python/AGENTS.mdAGENTS.md
- rust/AGENTS.mdAGENTS.md
- .github/copilot-instructions.mdCopilot instructions

The file

We split this CLAUDE.md into the rules it gives an agent

10 rules. About 0 are common, 10 are distinctive, and 0 are very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈10
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Distinctive:
NO task completion without tests covering it and passing
line 6
Distinctive:
NO PR creation without all subagent feedback addressed
line 7
Distinctive:
NEVER assume - always verify with tests and benchmarks
line 8
Distinctive:
NEVER ignore bugs, even out of scope - open an issue
line 9
Distinctive:
Ask if unsure, stop and reassess if looping
line 14
Distinctive:
If asked to do X a certain way, do it that way.
line 16
Distinctive:
Commit frequently with meaningful messages - git is our diary
line 19
Distinctive:
Keep PRs small and focused, split if too large
line 22
Distinctive:
Run tests for relevant scope
line 31
Distinctive:
Run subagents in parallel, address all feedback:
line 34

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.

Files

CLAUDE.md

Hard Constraints (non-negotiable)

NO task completion without tests covering it and passing
NO PR creation without all subagent feedback addressed
NEVER assume - always verify with tests and benchmarks
NEVER ignore bugs, even out of scope - open an issue

Rules

Always

Direct and concise, no compliments or apologies
Ask if unsure, stop and reassess if looping
Fetch web resources fresh, don't rely on cached data
If asked to do X a certain way, do it that way. Disagree? Raise it, but don't change without approval

When Writing

Commit frequently with meaningful messages - git is our diary
Focus on one language + its core bindings. Core changes = consider all bindings
AI agent files: context-efficient, use XML for structured data in Markdown
Keep PRs small and focused, split if too large

Before Task Completion

Tests cover it and pass
Linter/formatter ran for changed languages

Before Push

git pull --rebase upstream main, resolve conflicts
Exception: if on a feature branch rebasing onto another feature branch, rebase onto that branch instead
Run tests for relevant scope

Before PR Creation/Update

Run subagents in parallel, address all feedback:
- performance analysis
- code quality, style, best practices
- test coverage
- documentation clarity
- security vulnerabilities and edge cases

What the user cares about (all equally important):

Performance - low latency, high throughput
Reliability - robust error handling, edge cases
Usability - clear APIs, good documentation, best DX
Maintainability - clean code, modular design, tests, simplicity
Correctness - verify with tests and benchmarks, not assumptions

Project Structure

benchmarks/     # Performance benchmarks
docs/           # Documentation
examples/       # Usage examples for each language
ffi/            # Foreign function interface for language bindings
glide-core/     # Rust core driver - handles connection, protocol, clustering
glide-logger/    # Rust logging infrastructure
glide-rdma/     # Rust RDMA transfer layer over libfabric
go/             # Go client wrapper
java/           # Java client wrapper
node/           # Node.js client wrapper
python/         # Python client wrapper
utils/          # Test utilities and cluster management scripts

Architecture: Language Bindings to Core

Language	Mechanism	Native Library	Communication
Python Async	PyO3	valkey-glide (python/glide-async/)	Unix socket IPC
Python Sync	CFFI	glide-ffi (ffi/)	FFI calls
Java	JNI	glide-rs (java/)	JNI calls
Go	CGO	glide-ffi (ffi/)	FFI calls
Node.js	NAPI v2	rust-client (node/rust-client/)	Unix socket IPC

Socket IPC wrappers → socket_listener → glide-core → Valkey/Redis FFI wrappers → glide-core → Valkey/Redis

Context Retrieval

<context-sources>
  <language name="python-async">
    <triggers>python async, PyO3, glide-async, socket IPC python</triggers>
    <start-with>python/glide-async/src/lib.rs</start-with>
    <depends-on>core</depends-on>
    <entry>python/glide-async/Cargo.toml</entry>
    <bindings>python/glide-async/src/lib.rs</bindings>
    <client>python/glide-async/python/glide/glide_client.py</client>
    <commands>python/glide-shared/glide_shared/commands/</commands>
    <tests>python/tests/</tests>
  </language>
  <language name="python-sync">
    <triggers>python sync, CFFI, glide-sync, synchronous python</triggers>
    <start-with>python/glide-sync/glide_sync/_glide_ffi.py</start-with>
    <depends-on>ffi</depends-on>
    <entry>python/glide-sync/setup.py</entry>
    <bindings>python/glide-sync/glide_sync/_glide_ffi.py</bindings>
    <client>python/glide-sync/glide_sync/glide_client.py</client>
    <commands>python/glide-sync/glide_sync/sync_commands/</commands>
    <tests>python/tests/</tests>
  </language>
  <language name="java">
    <triggers>java, JNI, glide-rs, Java client</triggers>
    <start-with>java/src/lib.rs</start-with>
    <depends-on>core</depends-on>
    <entry>java/Cargo.toml</entry>
    <bindings>java/src/lib.rs</bindings>
    <client>java/client/src/main/java/glide/api/</client>
    <commands>java/client/src/main/java/glide/api/commands/</commands>
    <tests>java/client/src/test/java/glide/</tests>
  </language>
  <language name="node">
    <triggers>node, nodejs, NAPI, typescript, rust-client node</triggers>
    <start-with>node/rust-client/src/lib.rs</start-with>
    <depends-on>core</depends-on>
    <entry>node/rust-client/Cargo.toml</entry>
    <bindings>node/rust-client/src/lib.rs</bindings>
    <client>node/src/BaseClient.ts</client>
    <commands>node/src/Commands.ts</commands>
    <tests>node/tests/</tests>
  </language>
  <language name="go">
    <triggers>go, golang, CGO, go client</triggers>
    <start-with>go/base_client.go</start-with>
    <depends-on>ffi</depends-on>
    <entry>go/Makefile</entry>
    <bindings>go/callbacks.go</bindings>
    <client>go/base_client.go</client>
    <commands>go/internal/interfaces/</commands>
    <tests>go/integTest/</tests>
  </language>
  <language name="core">
    <triggers>glide-core, rust core, socket listener, protocol, clustering</triggers>
    <start-with>glide-core/src/client/mod.rs</start-with>
    <entry>glide-core/Cargo.toml</entry>
    <client>glide-core/src/client/mod.rs</client>
    <socket>glide-core/src/socket_listener.rs</socket>
    <protobuf>glide-core/src/protobuf/</protobuf>
    <tests>glide-core/tests/</tests>
  </language>
  <language name="ffi">
    <triggers>ffi, foreign function interface, C bindings, libglide</triggers>
    <start-with>ffi/src/lib.rs</start-with>
    <entry>ffi/Cargo.toml</entry>
    <bindings>ffi/src/lib.rs</bindings>
  </language>
  <language name="rdma">
    <triggers>rdma, libfabric, efa, glide-rdma</triggers>
    <start-with>glide-rdma/src/lib.rs</start-with>
    <entry>glide-rdma/Cargo.toml</entry>
    <bindings>glide-rdma/glide-libfabric-sys/src/lib.rs</bindings>
    <tests>glide-rdma/tests/</tests>
    <core-integration>glide-core/src/rdma/, glide-core/src/client/rdma.rs, glide-core/src/client/rdma_connection.rs</core-integration>
    <core-tests>glide-core/tests/test_rdma.rs, glide-core/tests/test_rdma_remote.rs</core-tests>
    <ffi-integration>ffi/src/rdma_ffi.rs</ffi-integration>
    <python-integration>python/glide-sync/glide_sync/glide_client.py, python/glide-shared/glide_shared/config.py</python-integration>
    <python-tests>python/tests/sync_tests/test_sync_rdma.py, python/tests/test_rdma_config.py</python-tests>
  </language>
</context-sources>

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

It worked
Partly
It didn't

What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free