Skip to content

CowboyLogic/ai-dev / skills/client-config-copilotcli

client-config-copilotcli

Unverified

Manage GitHub Copilot CLI configuration files including settings.json, config.json, permissions-config.json, mcp-config.json, lsp-config.json, hooks, skills, custom agents, plugins, and custom instructions. Use this skill whenever the user wants to view, edit, add, or understand any Copilot CLI setting: trusted folders, user or repository settings, tool permissions, sandboxing, MCP or LSP servers, hooks, skills, custom agents, plugins and marketplaces, BYOK models, authentication, session management, or custom instructions. Trigger on: "add trusted folder", "allow tool", "configure MCP", "add hook", "create skill", "create agent", "install plugin", "set model", "use my own API key", "add custom instructions", "name session", "resume session", or "show my copilot config".

FileKindSize
CHANGELOG.mdmarkdown19.4 KB
README.mdmarkdown2.3 KB
SKILL.mdskill md7.6 KB
references/agents-plugins.mdmarkdown11.5 KB
references/cli-commands.mdmarkdown15.2 KB
references/config-schema.mdmarkdown37.2 KB
references/hooks.mdmarkdown15.6 KB
references/instructions.mdmarkdown4.1 KB
references/mcp.mdmarkdown12.1 KB
references/permissions.mdmarkdown22.8 KB
references/skills.mdmarkdown5.8 KB
scripts/show-config.pyscript14.5 KB
scripts/update-references.pyscript5.6 KB
sources.jsonconfig12.2 KB

references/skills.md

View on GitHub
# Skills Reference

Skills are folders of instructions, scripts, and resources. Copilot injects a skill's `SKILL.md`
when the prompt matches its `description`, or when invoked as `/SKILL-NAME`. All files in the
skill's directory are available to the agent.

## Locations (priority order — first found wins for a duplicate name)

| Location | Scope |
|----------|-------|
| `.github/skills/<name>/SKILL.md` | Project |
| `.agents/skills/<name>/SKILL.md` | Project |
| `.claude/skills/<name>/SKILL.md` | Project (Claude-compatible) |
| Parent directories' `.github/skills/` | Inherited (monorepos) |
| `~/.copilot/skills/<name>/SKILL.md` | Personal |
| `~/.agents/skills/<name>/SKILL.md` | Personal |
| Plugin `skills/` directories | Plugin |
| `COPILOT_SKILLS_DIRS` (comma-separated) and the `skillDirectories` setting | Custom |
| `.github/skills/` under `--add-dir` / `/add-dir` / SDK `additionalDirectories` | Added root (a trust decision: runs with project-skill trust) |
| Bundled with the CLI | Built-in (lowest; overridable by any other source) |
| Org/enterprise remote skills | Remote — listed last in the docs, but not given a rank: the docs say remote skills follow the same name-based priority as local skills, so whether one overrides a built-in with the same name isn't stated. Projected via the AHP relay and fetched on invocation |

The `ignoredSkillsLocations` setting (`string[]`, `~` paths allowed) excludes directories and their
descendants from discovery regardless of which location would surface them.

`~/.claude/skills/` is not a documented location — only the project-level `.claude/skills/` is.
Skill directory names should be lowercase with hyphens. Two plugins with the same skill name
coexist as `/plugin-a/search` and `/plugin-b/search`; the bare name routes to the higher-priority
plugin (skills only — commands keep tier-based dedup).

**Commands** (alternative format): individual `.md` files in `.claude/commands/`; the filename is
the command name; they support `argument-hint`, `description`, `allowed-tools`, and
`disable-model-invocation`. Skills beat commands with the same name.

---

## SKILL.md

```markdown
---
name: image-convert
description: Converts SVG images to PNG. Use when asked to convert SVG files.
allowed-tools: shell
---

When asked to convert an SVG to PNG, run `convert-svg-to-png.sh` from this skill's base
directory, passing the input SVG path as the first argument.
```

### Frontmatter fields

| Field | Type | Required | Description |
|-------|------|----------|-------------|
| `name` | string | Yes | Starts with a letter or number; letters, numbers, hyphens, underscores, dots, colons, spaces; max 64 chars. Colons allow namespaced names (`my-plugin:search`). Lowercase-with-hyphens is the documented convention and usually matches the directory |
| `description` | string | Yes | What it does and when to use it; max 1024 chars; drives auto-activation |
| `license` | string | No | Description of the license that applies to the skill |
| `argument-hint` | string | No | Hint shown in the skill picker, e.g. `"[target] [mode]"` |
| `allowed-tools` | string or string[] | No | Tools auto-allowed while the skill is active (comma list or YAML array; `"*"` for all) |
| `user-invocable` | boolean | No | Allow `/SKILL-NAME` (default `true`) |
| `disable-model-invocation` | boolean | No | Prevent automatic invocation (default `false`) |

**v1.0.99 release notes only:** a rule in `allowed-tools` keeps commas that appear inside its
arguments instead of splitting there, and where a list's structure is ambiguous its unscoped rules
now prompt instead of being granted. Prefer a YAML array over a comma list for rules with arguments.

> [!WARNING]
> Only pre-approve `shell` or `bash` in `allowed-tools` for skills (and scripts) you have reviewed
> and trust. It removes the confirmation step for terminal commands.

---

## Using skills

- Automatic: Copilot matches the prompt against `description`.
- Explicit: `Use the /frontend-design skill to ...` or just `/frontend-design`.
- Disable without deleting: `disabledSkills` setting (user or repo), `/skills` toggle, or
  `copilot skill disable NAME`.
- `dynamicRetrieval: { "skills": false }` turns off embeddings-based retrieval for skills.

## Management

### In a session

| Command | Purpose |
|---------|---------|
| `/skills` | Plugins dashboard on the Skills tab (enable/disable) |
| `/skills list` | List available skills |
| `/skills info NAME` | Details, including file location and source plugin |
| `/skills add [--project] <FILE\|URL\|DIRECTORY>` | Add a skill; `--project` scopes a file/URL install to this repo |
| `/skills remove <NAME\|DIRECTORY>` | Remove a skill, or unregister a custom directory |
| `/skills reload` | Reload without restarting |

### In the terminal

```bash
copilot skill list [--json]                  # rows: { name, description, source, path, enabled }
copilot skill add ./my-skill/SKILL.md        # personal (default)
copilot skill add --project ./my-skill/SKILL.md   # into .github/skills (file or URL only)
copilot skill enable my-skill
copilot skill disable my-skill
copilot skill remove my-skill
```

- Adding a **directory** registers it as a custom skill source (no copy). Adding a **file or URL**
  copies it into the personal or project skills directory.
- Only personal/project skills you added can be removed; plugin and built-in skills can only be
  disabled.
- `gh skill` (GitHub CLI) can also search, install, update, and publish skills.
- The retired `copilot plugins install --skill` / `--skill` flags are replaced by `copilot skill`.

---

## Skills vs custom instructions

| | Skills | Custom instructions |
|---|--------|---------------------|
| Best for | Detailed, task-specific guidance | Rules relevant to almost every task |
| Loaded | When relevant, or `/skill-name` | Always |
| Can bundle scripts | Yes | No |
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall