Is my-claude-code-mods open source?

Yes — YeonwooSung/my-claude-code-mods is open source, released under the Apache-2.0 license.

What language is my-claude-code-mods written in?

YeonwooSung/my-claude-code-mods is primarily written in TypeScript.

How popular is my-claude-code-mods?

YeonwooSung/my-claude-code-mods has 1.1k stars on GitHub.

Where can I find my-claude-code-mods?

YeonwooSung/my-claude-code-mods is on GitHub at https://github.com/YeonwooSung/my-claude-code-mods.

← all repositories

YeonwooSung/my-claude-code-mods

Seventeen mods that make Claude Code show its receipts

A grab-bag of Claude Code plugins for profiling sessions, rewinding edits, and catching the agent lying to itself.

★1.1k stars TypeScript Coding Assistants

View on GitHub ↗

Collecting fresh signals — velocity needs a few days of history.

collecting data…

star history

What it does

This is a personal collection of Claude Code mods — installable plugins — covering observability, guardrails, and workflow hygiene. The heavy hitters: agent-profiler breaks down where a session’s time actually goes (model wait, tools, subagents, re-reads) and exports OTLP traces for Jaeger/Tempo; time-machine records every tool call on a branching timeline with shadow-git snapshots so you can see files at any step and find which edit broke a test; loop-detective flags repeated Bash failures and edit/retry cycles with evidence.

The interesting bit

The newer half of the collection is almost a checklist of agentic-coding failure modes: context-inspector audits what’s filling your context window and what compaction threw away, assumption-tracker keeps evidence-backed states for the agent’s guesses, and verify-board marks results PASS/FAIL/STALE/UNKNOWN — acknowledging that a green check from three edits ago means nothing. It’s glue code in the best sense: no new models, just instrumentation for the parts of Claude Code that are otherwise a black box.

Key highlights

time-machine gives per-step file snapshots, restore, branching, and re-run — a debugger timeline for agent edits
agent-profiler exports OTLP traces, so your Claude sessions can live in the same Jaeger instance as your services
scope-guard and blast-radius warn on out-of-scope paths and reverse-import impact before edits land
token-run is a fully local ASCII runner for request failures — because apparently we need that now
Bilingual READMEs (English/Korean) for most mods, and a genuinely thorough validation matrix across Claude Code versions

Caveats

The ten workflow mods currently return command text only — no live panels
Live UI and simultaneous-mod loading are explicitly unverified; unit tests don’t establish UI compatibility
Requires a recent Claude Code CLI (mods enabled by default from 2.1.287); time-machine also needs git on your PATH

Verdict

Worth a look for anyone running long Claude Code sessions who wants to know where the time, tokens, and context actually went. If you only use Claude for quick one-shot edits, most of this is instrumentation for problems you don’t have yet.

Frequently asked

What is YeonwooSung/my-claude-code-mods?: A grab-bag of Claude Code plugins for profiling sessions, rewinding edits, and catching the agent lying to itself.
Is my-claude-code-mods open source?: Yes — YeonwooSung/my-claude-code-mods is open source, released under the Apache-2.0 license.
What language is my-claude-code-mods written in?: YeonwooSung/my-claude-code-mods is primarily written in TypeScript.
How popular is my-claude-code-mods?: YeonwooSung/my-claude-code-mods has 1.1k stars on GitHub.
Where can I find my-claude-code-mods?: YeonwooSung/my-claude-code-mods is on GitHub at https://github.com/YeonwooSung/my-claude-code-mods.