跳到正文
cccskills
免费在 GitHub 公开

uninstall

Completely remove claude-ops plugin, all stored credentials, cached files, shell exports, and MCP registrations. Confirms each step before deletion.

中文简介正在准备中。当前显示原文说明。

集成工具SlackTelegram
查看安装方法

包含的文件(1)

  • SKILL.md6.8 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

OPS > UNINSTALL

You are running a complete uninstall of the claude-ops plugin. This removes everything the plugin and /ops:setup created. Every deletion step requires user confirmation via AskUserQuestion — never delete silently.


Step 1 — Confirm intent

Use AskUserQuestion:

This will completely remove claude-ops and all its data:
  - Plugin installation and marketplace registration
  - Stored preferences and project registry
  - Cached plugin files
  - Keychain credentials (Telegram, Slack tokens)
  - Shell profile exports (CLAUDE_PLUGIN_ROOT)
  - MCP server registrations (Telegram, Slack)

  [Uninstall everything]  [Cancel]

If cancelled, stop immediately.


Step 2 — Detect what exists

Scan for all claude-ops artifacts. Build a list of what's present:

# Preferences
PREFS_DIR="${CLAUDE_PLUGIN_DATA_DIR:-$HOME/.claude/plugins/data/ops-ops-marketplace}"
ls -la "$PREFS_DIR" 2>/dev/null

# Cache
CACHE_DIR="$HOME/.claude/plugins/cache/ops-marketplace"
ls -la "$CACHE_DIR" 2>/dev/null

# Keychain entries (macOS)
for key in telegram-api-id telegram-api-hash telegram-phone telegram-session slack-xoxc slack-xoxd; do
  security find-generic-password -s "$key" 2>/dev/null && echo "FOUND: $key"
done

# Shell profile exports
grep -l 'CLAUDE_PLUGIN_ROOT' ~/.zshrc ~/.bashrc ~/.zprofile ~/.bash_profile 2>/dev/null

# MCP registrations
grep -l 'telegram\|slack-mcp' ~/.claude.json 2>/dev/null

Print a summary of what was found, then proceed to delete each category.


Step 3 — Remove keychain credentials

For each found keychain entry, ask via AskUserQuestion:

Remove keychain credential: <key-name>?
  [Yes]  [Skip]

On Yes:

security delete-generic-password -s "<key-name>" 2>/dev/null

Step 4 — Remove preferences and cache

Ask via AskUserQuestion:

Remove plugin data?
  - Preferences: <prefs-dir>
  - Cache: <cache-dir>

  [Yes, delete both]  [Skip]

On Yes:

rm -rf "$PREFS_DIR"
rm -rf "$CACHE_DIR"

Step 5 — Clean shell profile

For each shell profile file that contains CLAUDE_PLUGIN_ROOT:

Ask via AskUserQuestion:

Remove CLAUDE_PLUGIN_ROOT export from <file>?
  [Yes]  [Skip]

On Yes, read the file and remove lines containing CLAUDE_PLUGIN_ROOT. Use sed or similar — do NOT rewrite the entire file. Only remove the specific export line.

sed -i '' '/CLAUDE_PLUGIN_ROOT/d' "<file>"

Step 6 — Remove MCP registrations

Check if ~/.claude.json contains MCP server entries added by the plugin (telegram, slack-mcp). For each:

Ask via AskUserQuestion:

Remove MCP server registration: <server-name> from ~/.claude.json?
  [Yes]  [Skip]

On Yes, use jq to remove the entry:

jq 'del(.mcpServers["<server-name>"])' ~/.claude.json > /tmp/claude-json-tmp && mv /tmp/claude-json-tmp ~/.claude.json

Step 7 — Remove plugin from Claude Code settings

The /plugin uninstall UI does not always clean up settings files. Directly remove all ops references from the JSON config files that Claude Code reads on startup.

7a — Remove from installed_plugins.json

INSTALLED="$HOME/.claude/plugins/installed_plugins.json"
if [ -f "$INSTALLED" ] && grep -q 'ops.*marketplace' "$INSTALLED"; then
  python3 -c "
import json, sys
with open('$INSTALLED') as f: data = json.load(f)
keys_to_remove = [k for k in data.get('plugins', {}) if 'ops-marketplace' in k or 'ops@ops' in k]
for k in keys_to_remove: del data['plugins'][k]
with open('$INSTALLED', 'w') as f: json.dump(data, f, indent=2)
print(f'Removed {len(keys_to_remove)} entries from installed_plugins.json')
"
else
  echo "No ops entries in installed_plugins.json"
fi

7b — Remove from known_marketplaces.json

KNOWN="$HOME/.claude/plugins/known_marketplaces.json"
if [ -f "$KNOWN" ] && grep -q 'ops-marketplace' "$KNOWN"; then
  python3 -c "
import json
with open('$KNOWN') as f: data = json.load(f)
keys_to_remove = [k for k in data if 'ops-marketplace' in k or 'ops' in k.lower()]
for k in keys_to_remove: del data[k]
with open('$KNOWN', 'w') as f: json.dump(data, f, indent=2)
print(f'Removed {len(keys_to_remove)} marketplace entries from known_marketplaces.json')
"
else
  echo "No ops marketplace in known_marketplaces.json"
fi

7c — Remove from settings.json and settings.local.json

Both ~/.claude/settings.json and ~/.claude/settings.local.json may contain enabledPlugins entries and stale permissions.allow entries referencing ops-marketplace paths.

for SETTINGS_FILE in "$HOME/.claude/settings.json" "$HOME/.claude/settings.local.json"; do
  [ -f "$SETTINGS_FILE" ] || continue
  if grep -q 'ops' "$SETTINGS_FILE"; then
    python3 -c "
import json, re
with open('$SETTINGS_FILE') as f: data = json.load(f)
# Remove ops from enabledPlugins
ep = data.get('enabledPlugins', {})
ops_keys = [k for k in ep if 'ops' in k.lower()]
for k in ops_keys: del ep[k]
# Remove ops-marketplace permission entries
perms = data.get('permissions', {})
if 'allow' in perms:
  before = len(perms['allow'])
  perms['allow'] = [p for p in perms['allow'] if 'ops-marketplace' not in p and 'claude-ops' not in p]
  removed = before - len(perms['allow'])
else:
  removed = 0
# Remove ops from extraKnownMarketplaces
ekm = data.get('extraKnownMarketplaces', {})
ekm_keys = [k for k in ekm if 'ops' in k.lower()]
for k in ekm_keys: del ekm[k]
with open('$SETTINGS_FILE', 'w') as f: json.dump(data, f, indent=2)
print(f'Cleaned $SETTINGS_FILE: {len(ops_keys)} plugin entries, {removed} permissions, {len(ekm_keys)} extra marketplaces')
"
  else
    echo "No ops references in $SETTINGS_FILE"
  fi
done

7d — Remove marketplace and cache directories

rm -rf "$HOME/.claude/plugins/marketplaces/ops-marketplace" 2>/dev/null
rm -rf "$HOME/.claude/plugins/cache/ops-marketplace" 2>/dev/null
rm -rf "$HOME/.claude/plugins/data/ops-inline" 2>/dev/null
echo "Removed marketplace, cache, and data directories"

Step 8 — Final confirmation

Print:

claude-ops has been completely removed.

  Keychain:    <N> credentials deleted
  Preferences: deleted
  Cache:       deleted
  Shell:       <N> exports removed
  MCP:         <N> servers removed
  Settings:    cleaned (installed_plugins, known_marketplaces, settings)
  Directories: removed (marketplace, cache, data)

Run `/reload-plugins` then `/plugin` in Claude Code to verify ops no longer appears.

To reinstall later:
  /reload-plugins
  /plugin marketplace add Lifecycle-Innovations-Limited/claude-ops
  /plugin install ops@lifecycle-innovations-limited-claude-ops
  /reload-plugins
  /ops:setup

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

Automate ActiveCampaign tasks via Rube MCP (Composio): manage contacts, tags, list subscriptions, automation enrollment, and tasks. Always search tools first for current schemas.

中文简介正在准备中。当前显示原文说明。

davepoon/buildwithclaude36522026年10月19日 更新

Pull the ads a competitor is running right now on Meta (Facebook/Instagram) and Google, then extract the hooks, offers, CTAs and landing pages worth modelling. Use when writing ad copy, planning creative tests, choosing a landing page angle, or when the user asks "what is X advertising", "what ads does X run", "find competitor ad hooks". Reads live Ad Library and Ads Transparency data through the Tracetify MCP; falls back to a manual Ad Library walkthrough when the connector is absent.

中文简介正在准备中。当前显示原文说明。

davepoon/buildwithclaude36522026年10月19日 更新

Analytics your AI agent can actually use. Track, analyze, run A/B experiments, and optimize across all your projects via CLI. Includes a growth playbook so your agent knows HOW to grow, not just what to track.

中文简介正在准备中。当前显示原文说明。

davepoon/buildwithclaude36522026年10月19日 更新

Teaches when to recall from long-term memory before acting and when to save durable decisions, corrections and failures afterwards. Use when a memory tool or MCP memory server is connected but the agent is not using it consistently, when the user complains that the assistant forgets preferences, conventions or past decisions between sessions, or when setting up persistent memory for a project. Works with any memory backend: a folder of Markdown files, a local MCP server, or a managed service.

中文简介正在准备中。当前显示原文说明。

davepoon/buildwithclaude36522026年10月19日 更新

Audit local AI coding-agent sessions with agenttrace for cost, tokens, tool failures, latency, anomalies, health, diffs, and CI gates.

中文简介正在准备中。当前显示原文说明。

davepoon/buildwithclaude36522026年10月19日 更新

Checks which private files (.env, keys, secrets) Claude Code and its subagents reached in this project, whether a value came into an agent's context, and which key to change, by running the open source agentwhy CLI over the session records Claude Code already keeps on disk. Use when the user asks whether the agent or a subagent read their .env or other secrets, which private files a session touched, whether a key leaked into the conversation, or which key to rotate. Runs locally and never prints a secret value.

中文简介正在准备中。当前显示原文说明。

davepoon/buildwithclaude36522026年10月19日 更新

查看 davepoon 的全部技能

报告此技能的问题