CtrlK
BlogDocsLog inGet started
Tessl Logo

copilot-coding-agent

GitHub Copilot Coding Agent automation. Apply the ai-copilot label to an issue → GitHub Actions auto-assigns Copilot via GraphQL → Copilot creates a Draft PR. One-click issue-to-PR pipeline.

Install with Tessl CLI

npx tessl i github:supercent-io/skills-template --skill copilot-coding-agent
What are skills?

88

4.00x

Quality

83%

Does it follow best practices?

Impact

96%

4.00x

Average score across 3 eval scenarios

SKILL.md
Review
Evals

Evaluation results

100%

83%

Set Up GitHub Copilot Automation for a Repository

Copilot workflow setup infrastructure

Criteria
Without context
With context

COPILOT_ASSIGN_TOKEN secret

0%

100%

assign-to-copilot workflow deployed

20%

100%

ai-copilot label created

0%

100%

Workflow labeled trigger

58%

100%

CI workflow for Copilot PRs

100%

100%

GraphQL-Features header present

0%

100%

issues_copilot_assignment_api_support flag

0%

100%

coding_agent_model_selection flag

0%

100%

replaceActorsForAssignable mutation

0%

100%

Without context: $0.5728 · 2m 19s · 26 turns · 31 in / 9,457 out tokens

With context: $0.5458 · 1m 59s · 26 turns · 146 in / 6,917 out tokens

100%

95%

Create a Script to Directly Assign Issues to Copilot

GraphQL direct assignment script

Criteria
Without context
With context

COPILOT_ASSIGN_TOKEN env var

0%

100%

GraphQL-Features header

0%

100%

issues_copilot_assignment_api_support flag

0%

100%

coding_agent_model_selection flag

0%

100%

replaceActorsForAssignable mutation

0%

100%

Issue number argument

100%

100%

Without context: $0.3288 · 1m 29s · 17 turns · 22 in / 5,577 out tokens

With context: $0.5292 · 2m 14s · 21 turns · 56 in / 8,625 out tokens

88%

38%

Clear the Tech Debt Backlog with Copilot Automation

Bulk backlog processing and PR workflow

Criteria
Without context
With context

gh issue list with json flag

100%

100%

jq number extraction

100%

80%

xargs label application

0%

0%

ai-copilot label name

0%

100%

Draft PR behavior documented

100%

100%

Manual approval requirement

57%

100%

Copilot PR listing command

0%

100%

COPILOT_ASSIGN_TOKEN repo scope

0%

100%

External contributor note

100%

100%

Without context: $0.2482 · 1m 17s · 12 turns · 17 in / 4,540 out tokens

With context: $0.4246 · 1m 36s · 22 turns · 316 in / 5,353 out tokens

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.