agentleFS
ItamarZand88ItamarZand88/CLI-Anything-WEB/.claude/skills/sync-check/SKILL.md

sync-check

Skill215 starsChanged 3 months agoMITView on GitHub

Walks the documentation dependency web after a CLI code change and reports which downstream files (skills, SOPs, READMEs, plugin references) are out of sync. Use after fixing a bug, adding a command, changing auth behavior, refactoring, or before committing — and when the user says "sync check", "update docs", "propagate changes", "did I forget to update", "what else needs updating", or "before I commit".

4 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

8 rules. 0 are common, 5 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    After any code change, verify all related files are synchronized.

    line 8
  2. Very specific:

    Run /sync-check after making changes.

    line 13
  3. Distinctive:

    Check all downstream files that depend on the change

    line 15
  4. Distinctive:

    Every change touches multiple files.

    line 21
  5. Distinctive:

    Verify it matches the current code behavior

    line 126
  6. Distinctive:

    If you fixed a bug, ask:

    line 131
  7. Distinctive:

    If YES, add it to the appropriate plugin reference file:

    line 135
  8. Very specific:

    After all updates, run one command per affected CLI with --json to confirm the output still matches what the skill documents.

    line 151

“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Sync Check

After any code change, verify all related files are synchronized. This catches the "fixed the bug but forgot to update the skill/SOP/plugin docs" problem.

How to Use

Run /sync-check after making changes. It will:

  1. Detect what changed (from git diff or conversation context)
  2. Check all downstream files that depend on the change
  3. Report what's out of sync
  4. Offer to fix each one

The Dependency Web

Every change touches multiple files. Here's the full map:

Code Change → Documentation Updates

CLI code changed (client.py, commands/*.py, auth.py)
  ├── .claude/skills/<app>-cli/SKILL.md     (commands, output fields, examples)
  ├── <app>/agent-harness/<APP>.md           (SOP: rpcid table, data models)
  ├── <app>/agent-harness/cli_web/<app>/README.md  (usage examples, deps)
  ├── <app>/agent-harness/setup.py           (dependencies)
  └── <app>_cli.py _print_repl_help()        (REPL help must list all commands)

Bug Fix → Plugin Lesson Updates

Bug found and fixed in any CLI
  ├── Was it a pattern that could repeat in future CLIs?
  │   YES → Add to the relevant plugin reference file:
  │   ├── auth-strategies.md "Known Pitfalls" table    (auth bugs)
  │   ├── google-batchexecute.md                       (RPC/batchexecute bugs)
  │   ├── ssr-patterns.md "HTML Scraping Pitfalls"     (scraping bugs)
  │   ├── protection-detection.md                      (anti-bot/WAF bugs)
  │   └── traffic-patterns.md                          (protocol detection bugs)
  │
  ├── Is it a universal lesson (applies to ALL CLIs)?
  │   YES → Add to HARNESS.md "Lessons Learned" cross-reference table
  │
  └── Does it change a critical convention?
      YES → Update CLAUDE.md "Critical Conventions" section

New Command Added → Full Update Chain

New command added to a CLI
  ├── .claude/skills/<app>-cli/SKILL.md     (add command + options + output fields)
  ├── <app>_cli.py _print_repl_help()        (add to REPL help)
  ├── <app>/agent-harness/<APP>.md           (add endpoint to API map)
  ├── README.md (repo root)                  (if it changes the CLI's capabilities)
  └── Run: cli-web-<app> <new-cmd> --json    (verify output matches skill docs)

Auth/Protection Change → Cross-Cutting Updates

Auth behavior changed
  ├── auth-strategies.md                     (patterns, pitfalls, code examples)
  ├── HARNESS.md lessons table               (cross-reference)
  ├── testing/SKILL.md                       (auth prereqs)
  ├── standards/SKILL.md                     (smoke test steps)
  └── CLAUDE.md                              (critical conventions)

Anti-bot/protection change (e.g., httpx → curl_cffi)
  ├── protection-detection.md                (strategy, detection)
  ├── <app>/agent-harness/<APP>.md           (site profile, HTTP client)
  ├── <app>/agent-harness/setup.py           (dependencies)
  ├── <app>/agent-harness/cli_web/<app>/README.md  (deps section)
  ├── CLAUDE.md generated CLIs table         (protocol column)
  └── README.md (repo root)                  (protocol column)

Sync Check Procedure

Step 1: Identify What Changed

# Check git diff for changed files
git diff --name-only HEAD
# Or if already committed:
git diff --name-only HEAD~1

Step 2: Classify the Change

Ask yourself:

  • Which CLI? Extract from file path (booking, futbin, notebooklm, etc.)
  • What type of change?
    • Bug fix → needs lesson propagation?
    • New command → needs skill + REPL + SOP update?
    • Auth change → cross-cutting update?
    • Dependency change → setup.py + README?
    • RPC/protocol change → SOP + batchexecute ref?

Step 3: Run Verification Commands

For the affected CLI:

# Verify commands match skill docs
cli-web-<app> --help
cli-web-<app> <subgroup> --help

# Verify output fields match skill docs
cli-web-<app> <primary-command> --json | head -20

# Verify auth works (if auth CLI)
cli-web-<app> auth status --json

Step 4: Check Each Downstream File

For each file in the dependency chain above:

  1. Read the file
  2. Find the section that relates to your change
  3. Verify it matches the current code behavior
  4. Update if stale

Step 5: Plugin Lesson Check (Bug Fixes Only)

If you fixed a bug, ask:

"Would this bug happen again if someone generated a new CLI tomorrow?"

If YES, add it to the appropriate plugin reference file:

Bug CategoryAdd To
Auth (cookies, login, tokens, refresh)auth-strategies.md Known Pitfalls
RPC (wrong IDs, wrong params, parsing)google-batchexecute.md
HTML scraping (selectors, noise, slugs)ssr-patterns.md HTML Scraping Pitfalls
Anti-bot (blocked, 401, 403, challenges)protection-detection.md
Output (raw data leak, wrong fields)testing/SKILL.md CLI Output Sanity Checks
Downloads (URLs, cookies, formats)auth-strategies.md Known Pitfalls
General (UTF-8, Windows, rate limits)HARNESS.md Lessons table + relevant ref

Format for Known Pitfalls: | Symptom | Cause | Fix |

Step 6: Verify with --json

After all updates, run one command per affected CLI with --json to confirm the output still matches what the skill documents.

Quick Reference: File Locations

WhatWhere
CLI skills.claude/skills/<app>-cli/SKILL.md
CLI SOPs<app>/agent-harness/<APP>.md
CLI READMEs<app>/agent-harness/cli_web/<app>/README.md
Plugin capture skillcli-anything-web-plugin/skills/capture/SKILL.md
Plugin methodology skillcli-anything-web-plugin/skills/methodology/SKILL.md
Plugin testing skillcli-anything-web-plugin/skills/testing/SKILL.md
Plugin standards skillcli-anything-web-plugin/skills/standards/SKILL.md
Auth strategies refcli-anything-web-plugin/skills/methodology/references/auth-strategies.md
Batchexecute refcli-anything-web-plugin/skills/methodology/references/google-batchexecute.md
SSR patterns refcli-anything-web-plugin/skills/methodology/references/ssr-patterns.md
Protection refcli-anything-web-plugin/skills/capture/references/protection-detection.md
HARNESS.mdcli-anything-web-plugin/HARNESS.md
CLAUDE.mdCLAUDE.md (repo root)
README.mdREADME.md (repo root)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.