Verify Claudence Changes End-to-End

VerifiedCaution

Build, relaunch, and observe the Claudence menu bar app to verify a change end-to-end.

Sby Skills Guide Bot
DevelopmentIntermediate
7/22/2026
Claude Code
#macos#swiftui#build-verify#xcodebuild#menu-bar-app

Recommended for

Automated summary

This skill walks through building, relaunching, and observing the Claudence macOS menu bar app (SwiftUI) so a change can be verified end-to-end.

Strengths

  • Separates a fast swift build/swift test loop from the full xcodebuild app bundle build
  • Gives the exact relaunch sequence using pkill followed by open from DerivedData
  • Uses persisted snapshot, history, and diagnostics files to confirm behavior without accessibility access
  • Flags two real pitfalls: zsh globbing with --include and the blocking keychain permission dialog

Limitations

  • Tightly coupled to the Claudence app and its file paths, so it is not reusable as-is elsewhere
  • Cannot open the popover programmatically, so UI-only changes require a manual user screenshot
  • Assumes the user already has a Debug instance running from DerivedData
When to use it

Use it when you need to confirm a Claudence change compiles, relaunches cleanly, and writes fresh data into its persisted files.

When not to use it

Avoid it for verifying an unrelated macOS app or for testing the visual interface, which here needs manual action and a screenshot.

Security analysis

Caution
Quality score85/100

The skill is a local macOS development/verification workflow: Swift/Xcode builds, restarting a debug app, and inspecting its own persisted JSON output. No obfuscated payloads, no curl|sh, no secret or token transmission, and it explicitly warns against scripting keychain access. The only risk surface is ordinary powerful tooling (pkill, xcodebuild, open) and broad glob paths, so it warrants caution rather than a safe rating.

Findings
  • •Uses process-terminating command (`pkill -x Claudence`) that kills a running app without confirmation or backup of unsaved state
  • •Broad glob path `~/Library/Developer/Xcode/DerivedData/Claudence-*/...` may match multiple build directories, so `open` could launch a stale bundle
  • •Reads app-local JSON files (snapshots, history, diagnostics) which may contain account/usage data; no exfiltration or network calls present, and keychain access is explicitly avoided
How this analysis is made

Examples

Vérifier un changement de fetch
I just changed the usage fetch logic in Claudence. Build it, relaunch the Debug app, and check the snapshot's fetchedAt to confirm a fresh live fetch happened.
Boucle rapide avant build complet
Run swift build and swift test to check my Claudence changes compile, then do the xcodebuild Debug build and swap in the new app.
Diagnostiquer une erreur au lancement
Claudence isn't updating after my last edit — relaunch it and read diagnostics.json and the snapshots folder to figure out what's failing.

name: verify description: Build, relaunch, and observe the Claudence menu bar app to verify a change end-to-end.

Verifying Claudence changes

Claudence is a macOS menu bar app (SwiftUI, Xcode project + SwiftPM package in the same repo).

Build

  • Unit-test/compile loop: swift build and swift test from the repo root (fast, seconds).
  • Real app bundle: xcodebuild -scheme Claudence -configuration Debug build (single scheme Claudence; output lands in ~/Library/Developer/Xcode/DerivedData/Claudence-*/Build/Products/Debug/Claudence.app).

Relaunch

The user usually has a Debug instance running from DerivedData. To swap in a new build:

pkill -x Claudence; sleep 1; open ~/Library/Developer/Xcode/DerivedData/Claudence-*/Build/Products/Debug/Claudence.app

Keychain session key (com.claudence.session / default) survives relaunch; the app fetches live usage within seconds of launch.

Observe (no accessibility access needed)

The popover can't be opened programmatically (osascript lacks assistive access), but the app persists everything it renders:

  • Current snapshot: ~/Library/Application Support/Claudence/Snapshots/current_snapshot.json (also last_known_good_snapshot.json; check fetchedAt to confirm a fresh live fetch happened after relaunch)
  • History: ~/Library/Application Support/Claudence/History/usage_history.json (appended on each live fetch)
  • Diagnostics/errors: ~/Library/Application Support/Claudence/diagnostics.json
  • Settings: ~/Library/Application Support/Claudence/settings.json (session key is NOT here, it's in the Keychain)

For UI-only changes, ask the user to click the menu bar icon and screenshot — programmatic capture is blocked without accessibility + screen-recording permissions.

Gotchas

  • grep --include=*.swift fails under zsh globbing; quote it or use -r ... --include "*.swift".
  • Reading the keychain item from a script triggers a blocking permission dialog — don't; let the app do the fetch and read its output files instead.

From the file to a first result

The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.

View the source and supporting files

This download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.

1. Download the file

Download SKILL.md

From the root of a test project, save the file as .skill-downloads/verify-claudence-changes-end-to-end-mrwak9ih/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.

Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.

mkdir -p '.skill-downloads/verify-claudence-changes-end-to-end-mrwak9ih' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/verify-claudence-changes-end-to-end-mrwak9ih/download' --output '.skill-downloads/verify-claudence-changes-end-to-end-mrwak9ih/SKILL.md.part' && mv '.skill-downloads/verify-claudence-changes-end-to-end-mrwak9ih/SKILL.md.part' '.skill-downloads/verify-claudence-changes-end-to-end-mrwak9ih/SKILL.md'

2. Check prerequisites

Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.

No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.

The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.

3. Load the skill

After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.

Claude Code
test -f '.skill-downloads/verify-claudence-changes-end-to-end-mrwak9ih/SKILL.md' && test -s '.skill-downloads/verify-claudence-changes-end-to-end-mrwak9ih/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/verify' && cp -R '.skill-downloads/verify-claudence-changes-end-to-end-mrwak9ih' '.claude/skills/verify'

OpenSkills

OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.

Official OpenSkills installation
openskills install './.skill-downloads/verify-claudence-changes-end-to-end-mrwak9ih'

openskills read 'verify-claudence-changes-end-to-end-mrwak9ih'

Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.

For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.

openskills sync

Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.

4. Try a first task

Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:

Read .skill-downloads/verify-claudence-changes-end-to-end-mrwak9ih/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.

Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.

Related skills