Copilot Cli
Delegate coding tasks to GitHub Copilot CLI in non-interactive mode with explicit models, least-privilege tool flags, and session resume or share.
Overview
Copilot CLI is an agent skill most often used in Build (also Ship → review, Operate → iterate) that delegates tasks to GitHub Copilot CLI with non-interactive prompts, model selection, and permission controls.
Install
npx skills add https://github.com/giuseppe-trisciuoglio/developer-kit --skill copilot-cliWhat is this skill?
- Non-interactive patterns: `copilot -p`, `--model`, and `--silent` for programmatic runs
- Least-privilege permission flags: `--allow-tool`, `--allow-all-tools`, paths, URLs; `--yolo` called out as highest risk
- Session resume via `--resume` and share-to-markdown via `--share` or explicit paths
- Operational guidance: English prompts, explicit models for reproducibility, review output before applying changes
- Documents 3 core non-interactive invocation patterns plus 5 permission-flag variants and 3 session-control commands
Adoption & trust: 871 installs on skills.sh; 271 GitHub stars; 1/3 security scanners passed (skills.sh audits).
What problem does it solve?
You want Copilot to run a bounded coding task from your agent without babysitting an interactive terminal or granting excessive tool access by default.
Who is it for?
Builders who already use GitHub Copilot CLI and need repeatable, flag-documented delegation from another agent with explicit models and scoped permissions.
Skip if: Teams that forbid secondary agents on code, or workflows that require full autonomous apply without human review of Copilot output.
When should I use this skill?
Users ask to hand work to Copilot, run Copilot CLI programmatically, control permissions, pick a model, resume sessions, or share CLI output.
What do I get? / Deliverables
Copilot returns assistant output in a controlled CLI session you can resume, share as markdown, and review before applying repository changes.
- Copilot CLI assistant response (optionally silent stdout only)
- Optional shared markdown session transcript
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Primary shelf is build → agent-tooling because the skill is about orchestrating another coding agent from your main agent workflow. Copilot CLI delegation, permission flags, and session controls are agent-runtime ergonomics, not product frontend or release QA.
Where it fits
Parent agent runs `copilot -p` with `--allow-tool 'shell(git)'` to generate a focused commit message and diff summary.
Export a Copilot session with `--share ./copilot-session.md` before merging agent-suggested changes.
Resume an interrupted fix session with `copilot --resume <session-id>` after a production hotfix context switch.
How it compares
Skill reference for GitHub’s Copilot CLI—not the same as in-IDE Copilot chat or a generic `gh` automation script.
Common Questions / FAQ
Who is copilot-cli for?
Solo and indie developers orchestrating GitHub Copilot CLI from Claude Code, Cursor, Codex, or similar when they want scripted prompts and permission boundaries.
When should I use copilot-cli?
Use it in Build → agent-tooling to offload a subtask, in Ship → review when sharing a session markdown for audit, or in Operate → iterate when resuming a long Copilot session with `--resume`.
Is copilot-cli safe to install?
Check the Security Audits panel on this Prism page; the skill documents high-risk `--yolo` and broad allow flags—default to least privilege and review output before merging.
SKILL.md
READMESKILL.md - Copilot Cli
# Copilot CLI Command Reference (Skill Support) Minimal reference used by the `copilot-cli` skill. ## Core non-interactive patterns ```bash # Programmatic prompt execution copilot -p "<english prompt>" # Programmatic prompt with explicit model copilot -p "<english prompt>" --model <model-name> # Output only assistant response copilot -p "<english prompt>" --silent ``` ## Permission flags ```bash # Single tool (preferred when possible) copilot -p "<prompt>" --allow-tool 'shell(git)' # All tools copilot -p "<prompt>" --allow-all-tools # All paths copilot -p "<prompt>" --allow-all-paths # All URLs copilot -p "<prompt>" --allow-all-urls # Full permissions (highest risk) copilot -p "<prompt>" --yolo ``` ## Session controls ```bash # Resume a session copilot --resume <session-id> # Share session to markdown copilot -p "<prompt>" --share # Share session to explicit path copilot -p "<prompt>" --share ./copilot-session.md ``` ## Operational guidance - Keep prompts in English for consistency and model performance. - Prefer explicit model selection when reproducibility is important. - Use least-privilege permission flags; avoid `--yolo` unless explicitly requested. - Always review output before applying code changes. --- name: copilot-cli description: Provides GitHub Copilot CLI task delegation in non-interactive mode with multi-model support (Claude, GPT, Gemini), permission controls, output sharing, and session resume. Use when users ask to hand work to Copilot, compare models, or run Copilot programmatically from Claude Code. allowed-tools: Bash, Read, Write --- # Copilot CLI Delegation Delegate selected tasks from Claude Code to GitHub Copilot CLI using non-interactive commands, explicit model selection, safe permission flags, and shareable outputs. ## Overview This skill standardizes delegation to GitHub Copilot CLI (`copilot`) for cases where a different model may be more suitable for a task. It covers: - Non-interactive execution with `-p` / `--prompt` - Model selection with `--model` - Permission control (`--allow-tool`, `--allow-all-tools`, `--allow-all-paths`, `--allow-all-urls`, `--yolo`) - Output capture with `--silent` - Session export with `--share` - Session resume with `--resume` Use this skill only when delegation to Copilot is explicitly requested or clearly beneficial. ## When to Use Use this skill when: - The user asks to delegate work to GitHub Copilot CLI - The user wants a specific model (for example GPT-5.x, Claude Sonnet/Opus/Haiku, Gemini) - The user asks for side-by-side model comparison on the same task - The user wants a reusable scripted Copilot invocation - The user wants Copilot session output exported to markdown for review Trigger phrases: - "ask copilot" - "delegate to copilot" - "run copilot cli" - "use copilot with gpt-5" - "use copilot with sonnet" - "use copilot with gemini" - "resume copilot session" ## Instructions ### 1) Verify prerequisites ```bash # CLI availability copilot --version # GitHub authentication status gh auth status ``` If `copilot` is unavailable, ask the user to install/setup GitHub Copilot CLI before proceeding. ### 2) Convert task request to English prompt All delegated prompts to Copilot CLI must be in English. - Keep prompts concrete and outcome-driven - Include file paths, constraints, expected output format, and acceptance criteria - Avoid ambiguous goals such as "improve this" Prompt template: ```text Task: <clear objective> Context: <project/module/files> Constraints: <do/don't constraints> Expected output: <format + depth> Validation: <tests/checks to run or explain> ``` ### 3) Choose model intentionally Pick a model based on task type and user preference. - Complex architecture, deep reasoning: prefer high-capacity models (for example Opus / GPT-5.2 class) - Balanced coding tasks: Sonnet-class model - Quick/low-cost iterations: Haiku-class or mini models - If user specifies a model, respect it Use exact model names available i