context-fundamentals

This skill should be used when the user asks to "understand context", "explain context windows", "design agent architecture", "debug context issues", "optimize context usage", or discusses context components, attention mechanics, progressive disclosure, or context budgeting. Provides foundational understanding of context engineering for AI agent systems.

1.07x

Quality

63%

Does it follow best practices?

Impact

100%

1.07x

Average score across 3 eval scenarios

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./bundled/skills/context-fundamentals/SKILL.md

The canonical home for this skill is context-fundamentals in muratcankoylan/Agent-Skills-for-Context-Engineering

Evaluation results

100%

14%

AI Customer Support Agent Design

System prompt architecture and tool definition design

Criteria

Baseline

With context

Section delimiters used

91%

100%

Background section present

37%

100%

Instructions section present

100%

Tool guidance section present

100%

Output description section present

66%

100%

Altitude: not too low

87%

100%

Altitude: not too high

100%

Tool descriptions: what it does

100%

Tool descriptions: when to use

100%

Tool descriptions: what it returns

50%

100%

Tool disambiguation

90%

100%

No arbitrary character limits

100%

Context Budget Tracker for Long-Running Agents

Context budget monitoring and compaction triggers

Criteria

Baseline

With context

Token approximation: 4 chars/token

100%

Warning threshold at 70-80%

100%

Critical threshold above warning

100%

Breakdown by component

100%

Tool output component tracked

100%

Configurable context limit

100%

Utilization percentage reported

100%

Status indicator returned

100%

Demo produces visible output

100%

No raw character count as token count

100%

Degradation design comment

100%

On-Demand Documentation Loader for an AI Agent

Progressive disclosure and observation masking

Criteria

Baseline

With context

Summaries only at init

100%

On-demand full load

100%

Observation masking implemented

80%

100%

Masking uses reference ID

50%

100%

Masking stores full content

100%

Naming convention validator

100%

Demo: initial state shown

100%

Demo: on-demand load shown

100%

Demo: masking demonstrated

100%

Does NOT pre-load all content

100%

Manifest identifiers used

100%

Naming validator rejects vague names

100%

Repository: foryourhealth111-pixel/Vibe-Skills
Path: bundled/skills/context-fundamentals/SKILL.md
Commit: f627ab5

Evaluated: 5 months ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

AI Customer Support Agent Design Context Budget Tracker for Long-Running Agents On-Demand Documentation Loader for an AI Agent

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.