stateofthehart

9 mods across 1 repository, 2 stars between them.

debug

02

stateofthehart/claude-code-config

Command Claude Code

You are debugging an issue where the cause is unknown. This is different from /bugfix — here you don't know what's wrong yet. Follow this sequence.

not rated 2 6mo ago A 0 tokens

implement

03

stateofthehart/claude-code-config

Command Claude Code

You are implementing a feature. Follow this sequence. Do NOT start writing code until Step 3.

not rated 2 6mo ago A 0 tokens

prune

05

stateofthehart/claude-code-config

Command Claude Code

Run the session pruner. If the user passed no arguments (i.e. $ARGUMENTS is empty), run.

not rated 2 6mo ago A 0 tokens

refactor

06

stateofthehart/claude-code-config

Command Claude Code

You are refactoring code. The goal is to improve structure WITHOUT changing behavior. Follow this sequence.

not rated 2 6mo ago A 0 tokens

review

07

stateofthehart/claude-code-config

Command Claude Code

You are reviewing code. Use a subagent with fresh eyes — do NOT review in the same context you wrote the code. If you wrote the code in this session, spawn an independent reviewer subagent.

not rated 2 6mo ago A 0 tokens

test

08

stateofthehart/claude-code-config

Command Claude Code

You are writing tests for existing code. The goal is to catch real bugs, not achieve coverage metrics.

not rated 2 6mo ago A 0 tokens

context-management

09

stateofthehart/claude-code-config

Skill Claude CodeCodex

This skill should be used when the conversation is getting long, context feels bloated, Claude mentions compaction or context limits, the user asks about "context", "pruning", "memory", "compaction", or when Claude notices tool outputs are very large. Also activate when resuming a session, starting a large task, or…

not rated 2 6mo ago A 86 tokens

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: