Intelligent Terminal is an experimental fork of Windows Terminal that adds native support for AI agents through the Agent Client Protocol. It is for Windows users who want to manage compatible agent command-line tools alongside terminal tabs, profiles, themes, shells, and keyboard shortcuts, and the catalogue entries provide instructions, skills, hooks, and a plugin for working with it.
Borrowing it
Nothing to install: this file belongs to microsoft/intelligent-terminal. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/microsoft/intelligent-terminal/main/.github/skills/performance-review/SKILL.mdgit clone --depth 1 https://github.com/microsoft/intelligent-terminalWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/microsoft/intelligent-terminal/performance-review)<a href="https://agentmods.dev/skills/microsoft/intelligent-terminal/performance-review"><img src="https://agentmods.dev/badge/skills/microsoft/intelligent-terminal/performance-review/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/microsoft/intelligent-terminal/performance-review"><img src="https://agentmods.dev/badge/skills/microsoft/intelligent-terminal/performance-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00066 | $0.01928 |
| Opus 5.5 | $0.00026 | $0.00771 |
| Sonnet 5.5 | $0.00013 | $0.00386 |
| Haiku 4.5 | $0.00007 | $0.00193 |
Grade A, and why
performance-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Performance Review
Investigate how a change affects typing, rendering, tab switching, WTA startup, session refresh and repeated work. Establish the affected scenario and cost before recommending an optimization.
Review context
Establish the baseline and reviewed changes from the caller's local diff, commit range, branch comparison or PR. Record the exact snapshot for mutable work so later edits cannot be mistaken for reviewed evidence.
Identify changed files, affected components and callers. File and line counts help estimate effort; execution frequency, data growth and subsystem risk determine which paths merit deeper investigation. Caller-supplied scope and analyzer output are discovery inputs, not proof of complete coverage.
Review produces Markdown using the results template below. The caller owns execution permissions, additional output formats, validation and delivery. Review-only is the default; authorized corrections follow the repair criteria.
Investigation
Inspect the complete patch and compare behavior with the baseline. Trace the changed functions into repeated lifecycle and event paths: renderer and TerminalCore, text buffer, VT parser/adapter, UI thread, tab/pane ownership, WTA process pooling, tasks, locks, channels, session/log enumeration, allocations, copies and notification amplification.
Apply each rule when its trigger appears in the changes or affected callers. Record the rule ID with its evidence.
| Rule | Trigger | Investigation | Intentional cases |
|---|---|---|---|
PERF-REPEATED-WORK |
A loop, lookup, copy or enumeration changes on a repeated path. | Compare work as input grows; establish caller frequency and actual bounds. Inspect nested scans and collections used only for count or membership. | Fixed small bounds, one-time startup, required snapshots and amortized work. |
PERF-UI-BLOCKING |
A UI callback or dispatcher path gains synchronous work. | Trace the executing thread, I/O, waits and traversal through callees; identify the stalled interaction and its blocking cost. | Already-background work, bounded cheap callbacks and required thread affinity. |
PERF-MVVM-AMPLIFICATION |
A property, binding, collection update or subscription changes. | Trace notifications to consumers; count expensive refreshes per logical change, including bulk updates and reentrancy. | Required individual notifications, cheap consumers and existing batching. |
PERF-ASYNC-CONTENTION |
Lock scope, awaits, channels or synchronous work in a task changes. | Establish what survives suspension, who contends and how ordering affects progress; verify actual scopes. | Async-aware locks, released guards, intentional backpressure and non-suspending operations. |
PERF-OWNER-LIFETIME |
Tabs, panes, tasks, subscriptions, helpers or pools have different lifetimes. | Trace repeated creation through teardown and cancellation; identify retained work or memory beyond its intended owner. | Resident pools, pre-warmed helpers, stash/restore and deliberately longer-lived bounded tasks. |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 161 lines · 66 tokens per session scan A 68b34857fde6
performance-review is a skill published in the GitHub repository microsoft/intelligent-terminal (2,074 stars, last pushed today), licensed MIT. It adds 66 tokens to every session and 1,928 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-17.
Other skills, from other repositories
crossreview
Run when the user invokes /crossreview (or /crossreview:setup to choose the reviewers) or asks for a cross-review, a quorum review or a second opinion from other agents or models: fan a code review (uncommitted work, a branch, a commit range, files, a plan or a document) out to several console code agents (Claude…
hqe
Comprehensive codebase health auditing, remediation, and verification skill based on the canonical HQE Protocol v5.0.0.
cn-check
Install and run the Continue CLI (cn) to execute AI agent checks on local code changes. Use when asked to "run checks", "lint with AI", "review my changes with cn", or set up Continue CI locally.
learn-from-sage
Detection-gap (miss) analysis for Code Review Sage. Learn from shipped fixes, acted-on human comments, and design outcomes to close reviewer blind spots. Inline during review stages a candidate; a human triggers a one-shot AI consolidation into the live ruleset.
pr-writing-review
Extract and analyze writing improvements from GitHub PR review comments. Use when asked to show review feedback, style changes, or editorial improvements from a GitHub pull request URL. Handles both explicit suggestions and plain text feedback. Produces structured output comparing original phrasing with reviewer…
git
Git and code hosting platform management — GitHub, GitLab, Codeberg, Forgejo. Zero barrier — install x-cmd and manage repos from the terminal. Use for "git", "github", "gitlab", "repo", "pull request", "code hosting".