implement

A command that carries out a written implementation plan in a software project. It coordinates code exploration, task tracking, implementation, and review across isolated work areas.

In plain words
What is it for?
Use it when you already have an implementation plan and want the coding work delegated, reviewed, and completed phase by phase.
Why use it?
It gives a planned change a defined execution process and includes checks for whether the work matches the plan.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/juan294/cc-rpi/implement
Clone the repo
git clone --depth 1 https://github.com/juan294/cc-rpi

Made for: Claude Code.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 618 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00618
Opus 5 $0.00000 $0.00309
Sonnet 5 $0.00000 $0.00124
Haiku 4.5 $0.00000 $0.00062

Measured 2d ago against content hash d01f2ce0ff6f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

implement scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/implement.md · 44 lines

What it actually says

Implement the plan at: $ARGUMENTS

Model tier: sonnet — Sonnet 5 (1M context) session. All subagents: model: "sonnet".

Process:

  1. Read the plan completely. Check for existing checkmarks.
  2. Spawn Explore agents (model: "sonnet") to gather relevant context.
  3. Create tasks (TaskCreate) to track progress.
  4. Enter a worktree for implementation (EnterWorktree tool). Research and planning happen against main -- implementation must be isolated to avoid conflicts with other agents or uncommitted work on the shared branch.
  5. Check if remaining phases are marked [batch-eligible] in the plan.
    • If ALL remaining phases are batch-eligible, suggest using /batch to execute them in parallel (one worktree per phase, each opens a PR).
    • If user agrees, hand off to /batch with the plan reference. Done.
    • If user declines or phases have dependencies, continue sequentially below.
  6. For the CURRENT phase only: a. Delegate implementation to subagents (up to 3, model: "sonnet"). b. When done, submit to a reviewer subagent (model: "sonnet") focused on PLAN COMPLIANCE: does the code match what the plan specified? Are all items addressed? c. If reviewer requests fixes, send back to implementer. d. Repeat until reviewer approves. e. Run /simplify on the changed files — this is Anthropic's native code quality pass (reuse, quality, efficiency). It spawns 3 specialized agents and applies fixes. f. Run ALL automated verification commands (tests, typecheck, lint, build). g. Update checkboxes in the plan file. h. If you departed from the plan, append the decision to docs/plans/<plan-name>-notes.md under ## Deviations, as: plan said / found / chose / why. Deviations only -- never narration, and no file at all if the phase had none. /validate reads it. Commit it with the phase. You are in a worktree: an uncommitted notes file is destroyed at teardown, before /validate ever sees it. .gitignore excludes docs/plans/* but re-includes *-notes.md precisely so these survive -- do not re-ignore them. Reference the plan by NAME in backticks, never as a markdown link: the plan itself stays untracked, so a link to it dangles in a clean checkout and fails CI's link check.
  7. STOP. Report results and wait for human confirmation.
  8. Do NOT proceed to the next phase without confirmation.

If plan doesn't match reality:

  • STOP and present: Expected vs Found vs Why it matters.
  • Ask how to proceed.
Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 44 lines · 0 tokens per session scan A d01f2ce0ff6f

Subscribe to this mod's changes

implement is a command published in the GitHub repository juan294/cc-rpi (5 stars, last pushed 13d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 618 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.