Run any Skill in Manus with one click

$pwd:

performance-analysis

Name: Performance Analysis
Author: voidful

// Use when investigating slow execution, high memory usage, or excessive token consumption. Systematic measurement before optimization.

Run Skill in Manus

$ git log --oneline --stat

stars:1

forks:0

updated:May 11, 2026 at 08:45

SKILL.md

readonly

name	performance-analysis
description	Use when investigating slow execution, high memory usage, or excessive token consumption. Systematic measurement before optimization.
version	1.0.0
author	Aixlarity
license	Apache-2.0
metadata	{"aixlarity":{"tags":["performance","optimization","profiling","tokens"],"related_skills":["systematic-debugging","code-review"]}}

Performance Analysis

Core Principle

Measure first, optimize second. Never optimize based on intuition. Profiling data decides what to fix.

Two Performance Dimensions in Agent Systems

1. Runtime Performance (Rust code)

Standard profiling applies:

# Compile with debug symbols for profiling
cargo build --release
# Use system profiler (Linux)
perf record ./target/release/aixlarity exec "task"
perf report

2. Token Performance (LLM usage)

This is unique to agent systems. Every token costs money and latency.

Metrics to track:

Prompt tokens per turn — Is the context bloated?
Completion tokens per turn — Is the model over-generating?
Tool calls per task — Are there wasted tool calls (reading irrelevant files)?
Total cost per task — Aixlarity tracks this automatically per session.

Common Token Waste Patterns

Context stuffing: Injecting too many files via @{...} when only one is needed.
Skill catalog bloat: Too many skills listed in the prompt. Use Progressive Disclosure.
History accumulation: Long conversations without auto-compact triggering.
Redundant tool calls: Agent reads the same file multiple times across turns.

When to Use

When a task takes noticeably more tokens than expected.
When auto-compact triggers frequently (sign of context pressure).
Before adding new context sources to the prompt assembly.

related-skills.json

same repository

architecture-review.md

from "voidful/Aixlarity"

Use when evaluating whether a proposed change fits the existing architecture. Prevents layer violations, dependency cycles, and accidental coupling.

2026-05-111

code-review.md

from "voidful/Aixlarity"

Use when the user wants a code review instead of implementation. Prioritizes correctness bugs, behavioral regressions, missing tests, and risky assumptions.

2026-05-111

documentation-review.md

from "voidful/Aixlarity"

Use when reviewing documentation for accuracy, completeness, and alignment with source code. Catches doc-code drift before it confuses readers.

2026-05-111

git-workflow.md

from "voidful/Aixlarity"

Use when making changes that should be committed. Enforces atomic commits, meaningful messages, and clean history.

2026-05-111

refactoring.md

from "voidful/Aixlarity"

Use when improving code structure without changing behavior. Ensures each refactoring step preserves all existing tests.

2026-05-111

security-audit.md

from "voidful/Aixlarity"

Use when reviewing code for security vulnerabilities. Covers prompt injection, path traversal, command injection, and agent-specific attack vectors.

2026-05-111

package.json

"author": "voidful"

"repository": "voidful/Aixlarity"

View GitHub Repository View Creator Repositories

$ install --global

$ download --local

Run Skill in Manus

$ useful --forSOC

Software DevelopersComputer and Mathematical Occupations15-1252L4

name	performance-analysis
description	Use when investigating slow execution, high memory usage, or excessive token consumption. Systematic measurement before optimization.
version	1.0.0
author	Aixlarity
license	Apache-2.0
metadata	{"aixlarity":{"tags":["performance","optimization","profiling","tokens"],"related_skills":["systematic-debugging","code-review"]}}

Performance Analysis

Core Principle

Measure first, optimize second. Never optimize based on intuition. Profiling data decides what to fix.

Two Performance Dimensions in Agent Systems

1. Runtime Performance (Rust code)

Standard profiling applies:

# Compile with debug symbols for profiling
cargo build --release
# Use system profiler (Linux)
perf record ./target/release/aixlarity exec "task"
perf report

2. Token Performance (LLM usage)

This is unique to agent systems. Every token costs money and latency.

Metrics to track:

Prompt tokens per turn — Is the context bloated?
Completion tokens per turn — Is the model over-generating?
Tool calls per task — Are there wasted tool calls (reading irrelevant files)?
Total cost per task — Aixlarity tracks this automatically per session.

Common Token Waste Patterns

Context stuffing: Injecting too many files via @{...} when only one is needed.
Skill catalog bloat: Too many skills listed in the prompt. Use Progressive Disclosure.
History accumulation: Long conversations without auto-compact triggering.
Redundant tool calls: Agent reads the same file multiple times across turns.

When to Use

When a task takes noticeably more tokens than expected.
When auto-compact triggers frequently (sign of context pressure).
Before adding new context sources to the prompt assembly.

performance-analysis

Performance Analysis

Core Principle

Two Performance Dimensions in Agent Systems

1. Runtime Performance (Rust code)

2. Token Performance (LLM usage)

Common Token Waste Patterns

When to Use

More from this repository

More from this repository

Performance Analysis

Core Principle

Two Performance Dimensions in Agent Systems

1. Runtime Performance (Rust code)

2. Token Performance (LLM usage)

Common Token Waste Patterns

When to Use