CtrlK
BlogDocsLog inGet started
Tessl Logo

reskill

Extracts procedural knowledge from agent charters into shared reusable skills to optimize context budget. Audits charters for step-by-step procedures and checklists, identifies duplicates across agents, and produces slimmed-down charters under 1.5KB. Use when agent charters are too large, consolidating duplicated procedures, or reducing token usage by converting inline instructions into shared skill references.

78

1.12x
Quality

67%

Does it follow best practices?

Impact

96%

1.12x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./.squad/skills/reskill/SKILL.md
SKILL.md
Quality
Evals
Security

Evaluation results

100%

15%

Modernize a Multi-Agent Squad's Charter Files

Skill extraction and charter reference replacement

Criteria
Without context
With context

Skill files in correct location

0%

100%

Procedures removed from charters

100%

100%

Skill references added to charters

100%

100%

Charter size under 1.5KB

100%

100%

Identity preserved

100%

100%

Core behavioral rules preserved

100%

100%

All three charters updated

100%

100%

Procedural threshold respected

100%

100%

Reskill report produced

100%

100%

90%

7%

Consolidate Shared Procedures Across a Growing Agent Squad

Duplicate detection and cross-agent consolidation

Criteria
Without context
With context

Duplicates identified

100%

100%

Agents with overlap named

100%

100%

Shared skill files created

53%

100%

Skill path convention followed

0%

0%

All affected charters updated

100%

100%

Skill references in charters

100%

100%

Identity and rules preserved

100%

100%

Charter size targets met

100%

100%

Consolidation summary produced

100%

100%

100%

11%

Measure Context Budget Savings After Charter Optimization

Metrics reporting and charter size targets

Criteria
Without context
With context

Per-agent bytes before

100%

100%

Per-agent bytes after

100%

100%

Per-agent skills created

100%

100%

Aggregate bytes saved

100%

100%

Aggregate skills created

100%

100%

Tokens saved calculated

100%

100%

Skill files at correct path

0%

100%

Procedures extracted from both charters

100%

100%

Charter references added

87%

100%

Identity and rules intact

100%

100%

Charter size targets

100%

100%

Repository
FritzAndFriends/BlazorWebFormsComponents
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

Table of Contents

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.