target
Skill0 starsChanged 6 months agoMITView on GitHub
Manage project goals and sub-tasks in a local TARGET.md file. Use when the user invokes /target to view, set, or update the current project's goals and broken-down sub-tasks. Creates TARGET.md in the project root if it doesn't exist.
In this repository3 other files it gives its agents.
- agent-teamSkill
- loadSkill
- saveSkill
The file
We split this skill into the rules it gives an agent
9 rules. About 0 are common, 5 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 12
Always read TARGET.md first (if it exists) before doing anything
- Distinctive: line 15
Keep the file human-readable markdown
- Distinctive: line 31
Use this exact markdown structure:
- Distinctive: line 62
If file exists: display the contents formatted, then ask "Would you like to add a goal, mark a task done, or update the file?"
- Distinctive: line 63
If file doesn't exist: say "No TARGET.md found.
- Very specific: line 67
Read existing TARGET.md (create if missing)
- Distinctive: line 70
After user provides sub-tasks, add them to the file under the goal
- Very specific: line 71
Update the _Last updated_ date
- Very specific: line 84
If confirmed, start fresh: ask for goals and sub-tasks interactively, then write a new TARGET.md
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Target — Project Goal Tracker
Manage the current project's goals and sub-tasks stored in TARGET.md at the project root.
How to behave
- Always read
TARGET.mdfirst (if it exists) before doing anything - The file lives at
<project-root>/TARGET.md— use the current working directory as the project root - Display the current content clearly after any operation
- Keep the file human-readable markdown
Invocation modes
Parse the user's argument after /target:
| Argument | Action |
| (none) | Show current TARGET.md, or prompt user to define the first goal if file doesn't exist |
add <goal text> | Append a new top-level goal (with empty sub-task list) |
done <task text> | Mark a matching task or sub-task as done ([x]) |
reset | Ask for confirmation, then delete and recreate TARGET.md from scratch |
| any other text | Treat as a new goal definition and ask the user if they want to replace or append |
TARGET.md format
Use this exact markdown structure:
# Project Targets
_Last updated: YYYY-MM-DD_
---
## Goal: <goal title>
> <optional one-line description>
### Sub-tasks
- [ ] <sub-task 1>
- [ ] <sub-task 2>
- [x] <completed sub-task>
---
## Goal: <another goal>
...Workflow
On /target (no argument)
- Read
TARGET.md - If file exists: display the contents formatted, then ask "Would you like to add a goal, mark a task done, or update the file?"
- If file doesn't exist: say "No TARGET.md found. What's the main goal for this project?" then after the user answers, create the file using the format above. Ask the user to break it down into sub-tasks.
On /target add <goal>
- Read existing
TARGET.md(create if missing) - Append the new goal section with an empty sub-task list
- Prompt: "What sub-tasks should we break this goal into?"
- After user provides sub-tasks, add them to the file under the goal
- Update the
_Last updated_date
On /target done <task>
- Read
TARGET.md - Find the closest matching task (fuzzy match on text)
- Change
- [ ]to- [x] - Update the
_Last updated_date - Show the updated section
On /target reset
- Confirm with user before proceeding
- If confirmed, start fresh: ask for goals and sub-tasks interactively, then write a new
TARGET.md
After every write
Show the user the updated TARGET.md content in a fenced markdown block so they can verify.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

