97
Your agent, on the shoulders of giants.
Early beta. Works on Claude Code, GitHub Copilot CLI, and OpenCode.
Install ·
What's inside ·
Does it work? ·
FAQ ·
Credits ·
Development
What this is
Skills your coding agent uses at the moment of decision: when it's about
to refactor, write a test, design an API, or commit. They draw on
Fowler on refactoring, 12-factor on configuration and deploys,
Ousterhout and Liskov on API design, Nygard on resilience patterns,
and others.
The project is named after 97 Things Every Programmer Should Know
(O'Reilly, ed. Kevlin Henney) and follows the same idea: one principle
at a time, applied when it matters. Every principle is attributed to
its original author. Principles are language-agnostic and durable.
Install
Works on Linux, macOS, and Windows. Node 18+.
Three supported harnesses. Pick the one you use.
Claude Code
/plugin marketplace add oribarilan/97
/plugin install 97@97-marketplace
Updates ship via the marketplace; run /plugin update 97 when a new version is available.
GitHub Copilot CLI
copilot plugin marketplace add oribarilan/97
copilot plugin install 97@97-marketplace
Updates ship via the marketplace; run copilot plugin update 97 when a new version is available.
OpenCode
Add to your OpenCode config file.
{
"plugin": [
"97@git+https://github.com/oribarilan/97.git"
]
}
Restart OpenCode. The plugin pulls the latest commit on each restart.
Advanced: pinned install
For reproducible behavior across sessions, pin to a specific tag (#vX.Y.Z):
{
"plugin": [
"97@git+https://github.com/oribarilan/97.git#v0.5.0"
]
}
What's inside
| Skill | When it applies |
|---|
using-97 | Always — entry skill that loads the trigger map |
before-you-refactor | Considering, evaluating, or performing a refactor, restructure, cross-file rename, or cleanup |
clean-code | Writing or reviewing functions, classes, naming, or ≥3 lines of non-trivial logic (includes Single Responsibility Principle (SRP), DRY, KISS, YAGNI) |
domain-modeling | Introducing, reviewing, or renaming a top-level type/table/domain concept, or deciding where state lives |
api-design | Designing or reviewing a public API, function signature, module boundary, exported type, or any contract other code depends on (includes LSP, SRP at the boundary) |
testing-discipline | Writing or reviewing tests, designing test data, naming a test, choosing what to assert, or writing test helpers/mocks/fixtures |
correctness-traps | Writing or reviewing error handling, floating-point math, concurrent code, remote calls, singletons, hot-path data structures, or high-volume log statements |
security-and-trust-boundaries | Writing or reviewing code that parses user input, builds SQL/shell commands, handles secrets/credentials, changes auth checks, deserializes untrusted data, or constructs paths/URLs from input |
observability | Writing or reviewing request handlers, RPCs, or background jobs for production; adding tracing, metrics, or structured-log calls; or making diagnosability decisions |
build-deploy-and-tooling | Writing, reviewing, or changing build scripts, CI config, deploy pipelines, repo setup, or evaluating a new tool/dependency |
self-review | About to commit, finish a task, open a PR, summarize work, or when asked for a review or hand-off of your own work |
working-with-users-and-team | Gathering or interpreting requirements, estimating effort, or communicating with stakeholders about what to build |
12 skills total: one entry skill plus 11 themed skills. Every principle
is cited and attributed in the skill's principles.md.
Giants
Every principle the agent applies is attributed to its original author.
Each skill ships a principles.md next to its SKILL.md with, per
principle: the author, a link to the source (CC-BY-3.0 essay, book
chapter, RFC, or specification as appropriate), our paraphrase, and
how the agent applies it.
The sources currently cited:
