See it in action
One of 4 test cases the skill fixed. Same model, same prompt, with and without the skill.
PromptI am creating a markdown goal file for running a marathon in 2024. Which exact folder path should this active goal file be saved under?
What was checked, without and with the skill| Checked | Without skill | With skill |
|---|
| Specifies the directory path ~/goals/active/ | Missed | Passed |
|---|
What's inside
The instructions your model reads when the skill loads.
SKILL.md121 linessingle file
Core Behavior
- User mentions an aspiration → help clarify and structure as goal
- Track progress without nagging → surface when relevant
- Regular reviews → quarterly and yearly reflection
- Create
~/goals/ as workspace
Goal vs Project vs Habit
- Goal: outcome you want (run a marathon, save €10k, learn Spanish)
- Project: defined end state, series of tasks (plan the wedding)
- Habit: recurring behavior (exercise 4x/week)
- Goals often spawn projects and habits to achieve them
When User States a Goal
- "What does success look like specifically?"
- "By when?" — deadline creates urgency
- "Why does this matter to you?" — motivation for hard days
- "What's the first small step?"
Goal File Structure
One file per goal: run-a-marathon.md
- What: specific outcome
- Why: motivation and meaning
- By when: target date
- Milestones: checkpoints along the way
- Current status: on track, behind, ahead
- Progress log: dated updates
Milestone Design
Break big goals into checkable milestones:
- Marathon: 5k → 10k → half marathon → full
- Save €10k: €2.5k per quarter
- Learn Spanish: A1 → A2 → B1
- Each milestone is a mini-celebration
Folder Structure
~/goals/
├── active/
│ ├── run-marathon-2024.md
│ └── save-10k.md
Track record
Benchmark
22 / 22
test cases passed, vs 18 / 22 without
Adoption
- Users
- 0
- Times used
- 0
Safety and source details
SkillSafety
✓ 3 of 3 passed✓Code scan15 of 15 checks clean
0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026
View 15 scanner checks
live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing
✓IntentSafe
LLM judge · does the skill do what it claims
✓ContentClean
LLM judge · policy & harmful-content review
Source
Importedberabuddies / semia
tests/fixtures/skills/ivangdavila/goals/SKILL.md
View on GitHub ↗License Apache-2.0Imported 8/3/2026Version v1