Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
code-yeongyu avatar

Pi Goal

  • 1 installs
  • 17 repo stars
  • Updated July 26, 2026
  • code-yeongyu/pi-goal

Persistent goal tracking for the pi agent: create, inspect, update, pause, resume, and complete long-running goals with a token budget.

About

A skill providing Codex-style persistent goal tracking for pi, with tools to create, inspect, update, pause, resume, and complete a long-running goal. A developer uses it only when explicitly setting up or continuing persistent goal tracking or an active goal exists.

  • Goal tools cover create, get, update, pause, resume, and complete
  • Goals carry an objective and a token budget

Pi Goal by the numbers

  • 1 all-time installs (skills.sh)
  • Ranked #14,102 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
  • Data as of Jul 29, 2026 (Skillselion catalog sync)
npx skills add https://github.com/code-yeongyu/pi-goal --skill pi-goal

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Installs1
repo stars17
Last updatedJuly 26, 2026
Repositorycode-yeongyu/pi-goal

What it does

Persistent goal tracking for the pi agent: create, inspect, update, pause, resume, and complete long-running goals with a token budget.

Files

SKILL.mdMarkdownGitHub ↗

pi-goal

Use goal tools only when the user explicitly wants persistent goal tracking or when an active goal already exists.

Tools

Create a goal:

create_goal({
	objective: "Ship the pi-goal extension",
	token_budget: 50000,
});

Inspect a goal:

get_goal({});

Update a goal:

update_goal({
	status: "complete",
});

update_goal only accepts complete. User-facing /goal commands control pause, resume, budget-limited, and clear transitions.

Completion Rule

Before marking a goal complete, audit the actual current state:

1. Restate the goal as concrete deliverables. 2. Map every explicit requirement to real evidence. 3. Inspect files, command output, test results, or repository state for each item. 4. Treat uncertainty as incomplete. 5. Call update_goal({ status: "complete" }) only when no required work remains.

Use budget-limited status when the reason to stop is budget exhaustion rather than completion.

Related skills

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.