mcpField
G
goal

goal

github.com/alekzonderAgent workflows0107 installsUpdated 2d ago
Repository

About goal

A skill for selecting and maintaining the daemon's active Goal, which attributes AI usage to a Native Task and its top-level root. It explains how to set a Goal through the tasks skill and `scripts/goal.sh set`, and notes that selection remains sticky until released.

Key features

  1. 01Select an active Goal
  2. 02Attribute AI usage to a Native Task
  3. 03Set a Goal with `scripts/goal.sh set`
  4. 04Keep selection across iterations

Use cases

  1. 01Assign an iteration to an active task
  2. 02Ensure AI usage is attributed to its task root

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/alekzonder/tariboy-store --skill goal

After installing, type/goalin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

evals/
evals.json
scripts/
goal.py
goal.sh

Details

From the project README

Goal

The daemon selects the Goal shown in runtime: goal and automatically attributes AI usage to that Native Task and its top-level root. Use the tasks skill to read, create, and update the authoritative task.

If this iteration began without a Goal, create or claim the task first, then run scripts/goal.sh set <TASK-KEY>. The task must be active and assigned to this agent. The command output confirms selection for subsequent AI requests; the already-rendered runtime: goal block does not refresh during the iteration, and earlier unattributed requests remain unchanged.

A selected Goal remains daemon-owned and sticky across iterations until released. Neither another goal.sh set call nor merely starting a new iteration can replace it.

Comments

Markdown supported