Click Path Audit
Finds UI buttons whose handler steps cancel each other out or leave the app in a wrong final state.
- Repository
- affaan-m/ECC
- Stars
- 295,127
- Licence
- MIT
- Files
- 1 file
What it does
When installed, the agent traces what happens when a user clicks a button, step by step through every state change, instead of only checking that each function exists and runs. It looks for cases where individually working functions undo each other through shared state, such as one action silently resetting a flag another action just set. It also builds a map of store actions and their side effects to make these hidden interactions visible.
Great for
- Developers debugging a button that appears to do nothing even though its functions seem to work
- Teams who just refactored a shared state store (Zustand, Redux, React context) and want to check for side effects
- Engineers who finished a normal debugging pass and suspect deeper state-interaction bugs remain
Skip it if
- Projects without interactive UI or shared state, because the audit has little to find there
- Non-developers looking for a general bug fixer; the skill assumes familiarity with React-style handlers and state stores
- Projects where the agent cannot read the relevant source files, since the audit depends on reading code
Try asking
- “Audit the 'New Email' button and trace every state change its click handler causes.”
- “Map the side effects of all actions in our Zustand emailStore and flag any that reset state they do not own.”
- “A user reports the Save button does nothing; check the full handler sequence for cancelled state changes.”
- “After refactoring our checkout state store, audit the touchpoints that read from it for conflicting final states.”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 1 original files, the MIT licence and an attribution note.
click-path-audit.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip click-path-audit.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip click-path-audit.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Name the specific component or button and the state store involved so the agent knows the scope of the audit.
- Ask for the store side-effect map first, then audit individual touchpoints against it, since the map is the key reference.
- Treat findings as hypotheses and confirm them by reproducing the click path in the running app or in tests.
What's inside
- SKILL.md
Redistributed unmodified from affaan-m/ECC under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.