Install
$ agentstack add skill-kevinzai-commander-ccc-more ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
About
/ccc-more — Second-page tools menu
Click-first picker for the CC Commander surface area that's NOT build/review/ship/design/learn. Plan, x-ray, Linear, fleet, connect, browse — one click each, cascade for the long tail.
Response shape (EVERY time)
1. Brand header (one line)
**CC Commander · More** · 7 tools · [all commands](https://commanderplugin.com/commands)
2. Context strip (one line)
Parallel Bash probe (silent on failure):
test -f tasks/todo.md && wc -l tasks/todo.md— open tasksgit rev-parse --abbrev-ref HEAD 2>/dev/null— branchclaude mcp list 2>/dev/null | grep -c '^ '— connected MCPstest -f .linear/config.json && echo linear-ok— Linear wired
Render: > 🧭 Branch ` · open tasks · MCPs connected · Linear: `
If barely any context: "🧭 Empty-ish project — Plan a feature or run X-ray to get signal."
3. The picker — AskUserQuestion page 1
Read ${CLAUDE_PLUGIN_ROOT}/menus/ccc-more.json once. Use its choices. Max 4 options — 3 headline tools + "Even more…".
question: "Which tool?"
header: "CC Commander · More"
multiSelect: false
options:
- label: "📋 Plan a feature"
description: "Spec interview → implementation plan file → review checkpoint."
preview: "Invokes ccc-plan. Asks 5-7 targeted questions — writes to the session plan file (shows in Desktop Plan pane)."
- label: "🩻 Project x-ray"
description: "Scorecard across code, docs, tests, deps, CI — actionable findings."
preview: "Invokes ccc-xray. Reads repo, returns report card + remediation steps."
- label: "📊 Linear board"
description: "View, pick, or create issues without leaving Claude."
preview: "Invokes ccc-linear. Requires Linear MCP; will prompt /ccc-connect if missing."
- label: "⋯ Even more…"
description: "Fleet, connect apps, browse catalog."
preview: "3 more tools — click to cascade."
Recommendation logic (prepend ⭐ to ONE option):
- No
tasks/dir and noCLAUDE.md→ ⭐ "Plan a feature" - Repo >500 files or messy tests → ⭐ "Project x-ray"
- Linear MCP connected + current branch matches
cc--*→ ⭐ "Linear board" - Nothing special → ⭐ "Plan a feature"
4. Handle the selection (page 1)
Dispatch immediately — no re-prompting:
- Plan a feature → invoke
ccc-planskill. - Project x-ray → invoke
ccc-xrayskill. - Linear board → check MCP list for
linear. If missing, suggest/ccc-connect linearfirst. If present, invokeccc-linear. - Even more… → cascade to page 2 below.
5. Second AUQ (only if "Even more…" picked)
question: "The rest."
header: "CC Commander · More · Page 2"
multiSelect: false
options:
- label: "🚁 Fleet (parallel agents)"
description: "Run multiple Sonnet agents in parallel worktrees — watch the progress cards."
preview: "Invokes ccc-fleet. Spawns worktree per agent, non-overlapping file domains."
- label: "🔌 Connect apps"
description: "Enable Notion, Zapier, Supabase, Slack, Google Drive, and more MCPs."
preview: "Invokes ccc-connect. Browses connector categories + runs OAuth."
- label: "🔎 Browse catalog"
description: "Searchable grid of every /ccc-* command and every installed skill."
preview: "Invokes ccc-browse. Filters by domain, status, recency."
- label: "🏠 Back to main"
description: "Return to /ccc."
preview: "Invokes ccc skill (root picker)."
6. Handle the selection (page 2)
- Fleet → invoke
ccc-fleetskill. - Connect apps → invoke
ccc-connectskill. - Browse catalog → invoke
ccc-browseskill. - Back to main → invoke
cccskill.
7. Argument handling
If the user passed an argument (/ccc-more fleet), skip both AUQs and invoke directly. Accept: plan / xray / linear / fleet / connect / browse.
Anti-patterns — DO NOT
- ❌ Render all 7 options in one AUQ (max 4)
- ❌ Show a numbered list and ask the user to type a number
- ❌ Invoke Linear without checking MCP presence — prompts fail silently
- ❌ Skip the cascade — page 2 is how the long tail surfaces
- ❌ Redundantly re-introduce CC Commander — the user already clicked through from /ccc
When to invoke this skill
- user: what else can CCC do?
- assistant: Loads ccc-more, renders the 4-option picker with ⭐ on the contextual best pick.
- user: plan out this feature before I build it
- assistant: Loads ccc-more with
planargument, skips picker, invokes ccc-plan directly.
- user: I need parallel agents for this refactor
- assistant: Loads ccc-more with
fleetargument, skips both AUQs, invokes ccc-fleet directly.
Bottom line: 4 options on page 1, cascade to 3 more on page 2. Arguments bypass everything. Every pick dispatches inline to the named skill — no intermediate text menus.
> ⚙️ Fable contract: plan before build · verifier ≠ worker · prove before alarm · loops need gates · leave durable state — rules/fable-method.md
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: KevinZai
- Source: KevinZai/commander
- License: MIT
- Homepage: https://commanderplugin.com
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.