hjemmesidekongen/ai/tier-assignment
Trust tier 3
- community
Source check pending
Checking that the SKILL.md file is still in the repository.
Evaluates decomposed subtasks using a weighted 5-factor risk matrix and assigns model tiers (junior/senior/principal). Produces per-subtask justifications to support dispatch decisions and cost optimization.
Install
npx vskill@latest install hjemmesidekongen/ai/tier-assignmentSKILL.md
Tier Assignment
Maps each decomposed subtask to a model tier using a weighted 5-factor risk score.
Risk Matrix
| Factor | Weight | What it measures | |
|---|---|---|---|
| Scope | 1.0 | How many files/modules touched | |
| Reversibility | 0.8 | How hard to undo if wrong | |
| Ambiguity | 1.2 | How unclear the requirements are | |
| Impact | 1.0 | Blast radius of a mistake | |
| Dependencies | 1.0 | Coupling to other subtasks |
Each factor scored 1-5. Weighted sum determines tier:
| Score Range | Tier | Model | |
|---|---|---|---|
| 5.0 - 10.0 | junior | haiku | |
| 10.1 - 18.0 | senior | sonnet | |
| 18.1 - 25.0 | principal | opus |
Rules
- Tier is a recommendation, not a constraint. Dispatcher may escalate on failure.
- If a subtask touches security-sensitive files (auth, crypto, secrets), floor the tier at senior.
- If a subtask has 3+ downstream dependents, floor the tier at senior (high-impact node).
- Justification must reference the dominant risk factor driving the assignment.
- Attach the tier + justification directly to the subtask in the decomposition YAML.
Override Triggers
Some patterns auto-escalate regardless of score:
- Database migrations: principal
- Public API changes: senior minimum
- Deletion of shared modules: principal
See references/process.md for scoring algorithm, examples, and edge cases.
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingplugins/smedjen/skills/tier-assignment/SKILL.md
Popularity
- Repository stars
- 1
- 7-day trend score
- 8
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/hjemmesidekongen/ai/tier-assignment)