wise-implement-plan-auto

wise-implement-plan-auto is a skill for Claude Code, Codex from e1024kb/wise-claude. It costs 147 tokens per session (913 once invoked), scanned A, original, MIT.

An automated plan executor that reads a PLAN.md file, gives each task to a separate software agent, and processes task groups in phases.

In plain words
What is it for?
Use it to implement a written coding plan in a Git working tree, including plans created by ticket-planning workflows.
Why use it?
It removes the need to carry out a long implementation plan manually and keeps each task isolated. It also commits and verifies tasks as it goes.

Skill for Claude CodeCodex

Installs and runs on its own, but its text points at files inside its plugin — anything it tells you to read at a ${CLAUDE_PLUGIN_ROOT} path is only there once the plugin is installed. Installing the plugin gets both.

Part of the wise plugin — 32 skills, 22 agents, 1 hook shipped together

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 skills/e1024kb/wise-claude/wise-implement-plan-auto
Any agent
npx skills add e1024kb/wise-claude --skill wise-implement-plan-auto
Clone the repo
git clone --depth 1 https://github.com/e1024kb/wise-claude

Made for: Claude Code, Codex.

Or install wise, the plugin that ships this one along with the rest of its 32 skills, 22 agents, 1 hook.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for wise-implement-plan-auto

README.md
[![agentmods](https://agentmods.dev/badge/skills/e1024kb/wise-claude/wise-implement-plan-auto.svg)](https://agentmods.dev/skills/e1024kb/wise-claude/wise-implement-plan-auto)
Your own site
<a href="https://agentmods.dev/skills/e1024kb/wise-claude/wise-implement-plan-auto"><img src="https://agentmods.dev/badge/skills/e1024kb/wise-claude/wise-implement-plan-auto.svg" alt="Measured on agentmods" height="20"></a>
Per session 147 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 913 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 $0.00147 $0.00913
Opus 5 $0.00073 $0.00456
Sonnet 5 $0.00029 $0.00183
Haiku 4.5 $0.00015 $0.00091

Measured 4d ago against content hash fdca4ed5c739, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

wise-implement-plan-auto 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 4d 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.

plugins/wise/skills/wise-implement-plan-auto/SKILL.md · 79 lines

How it starts

The opening of the file, as written. The whole thing — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/wise-implement-plan-auto — execute a plan, autonomously

Why this skill exists

ticket-plan and ticket-auto produce a PLAN-*.md but no skill executes one. /wise-implement-plan-auto is that executor — a phase-gated model: parse the plan's task waves, dispatch one fresh-context executor agent per task in parallel, commit each task atomically, verify as it goes. It is the reusable building block the ticket-auto workflow's implement phase follows.

Arguments

Read $ARGUMENTS. The first whitespace-separated token, if present, is the path to the PLAN-*.md to implement. When absent, look for a single PLAN-*.md at the git toplevel and use it; if there are zero or several, stop and ask the user to name one.

Procedure

1. Resolve the worktree + plan

git rev-parse --show-toplevel

Use the toplevel as worktree. Resolve plan_path from $ARGUMENTS (or the discovery rule above).

2. Follow the shared fragment

Read ${CLAUDE_PLUGIN_ROOT}/workflows/ticket-auto/prompts/implement-plan.md and follow it end to end with plan_path, worktree, project.kind (infer from the worktree's manifest if unknown), and SUPERVISE=yes. The fragment processes waves in order, dispatches the wave's executors per task (persona: this skill's agents/executor.md) — supervised background teammates a leader loop nudges if one hangs or goes idle mid-task — then simplifies (per-task, scoped to the task's files via references/simplify-pass.md) and commits each task sequentially, and verifies per task. (To fall back to plain blocking Task executors, pass SUPERVISE=no / set WISE_WORKER_* env to tune the watchdog.)

3. Relay the result

The fragment's final line is IMPLEMENT: waves=<w> tasks=<t> done=<d> failed=<f>. Summarise for the user — waves run, tasks done, tasks failed (with which ones) — and remind them nothing was pushed (commit/push is a separate step).

Guardrails

  • Never call AskUserQuestion mid-run — the only prompt is the argument-resolution stop in §Arguments when the plan is ambiguous.
  • One atomic commit per task; never bundle tasks.
  • Executors edit files; only this skill simplifies (per-task, scoped) and commits, serially — never let parallel subagents race the git index. The heavier high-depth code-review branch gate is a separate, later pipeline step, not this skill's job.
  • Never git push — that is the caller's step.
  • A failed task does not abort the run.
  • Never invoke another wise action skill.

Read the full file on GitHub · 79 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 4d ago First seen · 79 lines · 147 tokens per session scan A fdca4ed5c739

Subscribe to this mod's changes

wise-implement-plan-auto is a skill published in the GitHub repository e1024kb/wise-claude (4 stars, last pushed 9d ago), licensed MIT. It adds 147 tokens to every session and 913 once invoked, about $0.0007 per session on Opus 5. 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.