token-usage-prediction Skill
Estimate and present the token-usage budget for an upgrade before execution. Run this when the user directly requests an estimate, or to present a budget that the assessment tool already appended to its output. Never auto-run it after assessment — that path is now handled programmatically by the assessment tool. Published by microsoft in upgrade-agent-plugins.
What is token-usage-prediction Skill?
Estimate and present the token-usage budget for an upgrade before execution. Run this when the user directly requests an estimate, or to present a budget that the assessment tool already appended to its output. Never auto-run it after assessment — that path is now handled programmatically by the assessment tool. Published by microsoft in upgrade-agent-plugins. This profile combines repository metadata with install, compatibility, and usage signals so developers can quickly decide whether it fits their agent workflow before opening the source repository.
Automated repository signals based on public metadata such as recency, license, installation evidence, and adoption. These are not a security audit or endorsement.
Key capabilities
- Includes SKILL.md support
- Reusable instructions support
- Developers using upgrade-agent-plugins
Technical details
- Install or run with Copy skill directory
When to use token-usage-prediction Skill
- Use it for developers using upgrade-agent-plugins.
Built with
Editorial notes
Source
- Creator: microsoft
- Repository: microsoft/upgrade-agent-plugins
- Skill file: plugins/upgrade-agent/upgrade/skills/system/token-usage-prediction/SKILL.md
What it does
Estimate and present the token-usage budget for an upgrade before execution. Run this when the user directly requests an estimate, or to present a budget that the assessment tool already appended to its output. Never auto-run it after assessment — that path is now handled programmatically by the assessment tool.
Skill instructions
Token Usage Prediction Pre-execution token budget for the active scenario. Lets the user decide whether to proceed, narrow scope, or switch models before the agent starts spending tokens. The prediction is token-only — no monetary cost is reported. When to call Token estimation is user-initiated. Only run it when one of these is true: 1. The user directly asks ("how big will this be?", "recompute estimate", "how many tokens will the plan take?", "estimate for task X") — honor the request regardless of scenario, even when no scenario is active. 2. You are presenting a budget the assessment tool already appended. The dotnet-version-upgrade assessment tool computes the post-assessment budget itself (gated by the UPGRADEAGENTESTIMATETOKENSAFTERASSESSMENT env var) and appends it — already formatted with presentation guidance — to its own output. When that block is present, just present it as-is; do not call predicttokenusage to recompute it. When that block is absent, do nothing: never call
Explore related resources
Frequently asked questions
What is token-usage-prediction?
token-usage-prediction is a open-source AI agent skill with Copy skill directory. Estimate and present the token-usage budget for an upgrade before execution.
Who is token-usage-prediction best for?
token-usage-prediction is best for reusing agent instructions, scripts, and references.
How do I install token-usage-prediction?
Install or run token-usage-prediction using Copy skill directory. Check token-usage-prediction for the latest setup command.
Is token-usage-prediction actively maintained?
token-usage-prediction may need a closer maintenance check before production use.
Auto-fetched from GitHub.