Agenfk Close — Productivity skill for Claude Code
Summarize implementation and finalise the task.
How to install Agenfk Close
Installs to ~/.claude/skills/cglab-public-agenfk-agenfk-close/SKILL.md
mkdir -p ~/.claude/skills/cglab-public-agenfk-agenfk-close && curl -fsSL https://raw.githubusercontent.com/cglab-public/agenfk/HEAD/commands/agenfk-close.md -o ~/.claude/skills/cglab-public-agenfk-agenfk-close/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Agenfk Close does
description: Summarize implementation and finalise the task
Use the `agenfk` CLI for all workflow operations (CLI-only is the default; read with `--json` for machine-readable output). If `mcp__agenfk__*` tools are present (installed with `--with-mcp`), the equivalent MCP tool is interchangeable.
You are executing the `/agenfk-close ` command as a **Closing Agent**. Follow these steps precisely:
**Step 1 — Collate History**
- Read the item details using
agenfk get --json.
Alternatives in Productivity
- Epic Decompose — Break an epic into task children without creating task branches 243.5k ★
- Worktree Status Check — Verify the current worktree environment and show task details 23.4k ★
- P7 — PUA P7 骨干模式 — 方案驱动执行 19.5k ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Agenfk Deep
Run a task using full Multi-Agent Orchestration (Deep Mode)
Agenfk Pause
Pause work on the current task, saving full context for later resumption
Summarize Roadmap
Summarize the backlog as a one-line-per-task roadmap, grouped by epic
Ldd Trace
Show the current LDD trace for this task, and summarize the last 5 entries from .ldd/trace.log if it exists in
Fire
Start a task the orchestrator way — clarify it, write the ledger, hand it out, verify the close
Tm
Task Master - plan, start, review, and close
Related Agents
Context Isolator
Runs data-heavy read-and-summarize operations in an isolated context window to prevent bloating the main conve
Issue Feature Review
Review GitHub feature requests with constructive skepticism. Summarize the ask, evaluate user impact and imple
Task Verifier
Internal agent for verifying task completion. Spawned automatically on task close. Do not invoke directly.