Skill Management
- 15 installs
- 52 repo stars
- Updated March 4, 2026
- bfollington/terma
skill-management is a Claude Code skill that is a quick reference for creating, updating, and managing SKILL.md files and slash commands.
About
skill-management is a Claude Code skill that serves as a quick reference for creating, updating, and managing skills. It documents skill locations, SKILL.md frontmatter fields, string substitutions, invocation control, subagent execution, and permissions. A developer uses it when working with SKILL.md files, slash commands, or extending Claude's capabilities.
- Quick reference for creating, updating, and managing Claude Code skills
- Documents SKILL.md frontmatter, locations, and invocation control
- Covers subagent execution, dynamic context injection, and permissions
Skill Management by the numbers
- 15 all-time installs (skills.sh)
- Ranked #482 of 782 Skill Development skills by installs in the Skillselion catalog
- Data as of Jul 28, 2026 (Skillselion catalog sync)
skill-management capabilities & compatibility
- Capabilities
- documentation · orchestration
- Use cases
- documentation · orchestration
What skill-management says it does
Create, update, and manage Claude Code skills. Use when working with SKILL.md files, slash commands, or extending Claude's capabilities.
Keep SKILL.md under 500 lines; move detailed reference to separate files.
Priority: enterprise > personal > project. Plugin skills use `plugin:skill` namespace.
npx skills add https://github.com/bfollington/terma --skill skill-managementAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 15 |
|---|---|
| repo stars | ★ 52 |
| Last updated | March 4, 2026 |
| Repository | bfollington/terma ↗ |
What it does
Reference SKILL.md structure and options when creating or managing Claude Code skills.
Who is it for?
Looking up SKILL.md frontmatter, invocation control, and skill locations.
When should I use this skill?
You are working with SKILL.md files, slash commands, or extending Claude's capabilities.
What you get
Correctly structured SKILL.md files with appropriate frontmatter, invocation control, and permissions.
- SKILL.md files
- slash commands
By the numbers
- Keep SKILL.md under 500 lines
- Skill budget is 2% of the context window
Files
Skills Quick Reference
Full docs: https://code.claude.com/docs/en/skills.md
Skill Locations
| Location | Path | Scope |
|---|---|---|
| Personal | ~/.claude/skills/<name>/SKILL.md | All your projects |
| Project | .claude/skills/<name>/SKILL.md | This project only |
| Plugin | <plugin>/skills/<name>/SKILL.md | Where enabled |
| Enterprise | Managed settings | All org users |
Priority: enterprise > personal > project. Plugin skills use plugin:skill namespace.
Directory Structure
my-skill/
├── SKILL.md # Required - main instructions
├── reference.md # Optional - detailed docs (loaded on demand)
├── examples/ # Optional - example outputs
└── scripts/ # Optional - executable scriptsKeep SKILL.md under 500 lines; move detailed reference to separate files.
Frontmatter Reference
---
name: skill-name # Display name (defaults to directory name)
description: What it does # RECOMMENDED - helps Claude decide when to use
argument-hint: [issue-number] # Shown in autocomplete
disable-model-invocation: true # Only user can invoke via /name
user-invocable: false # Only Claude can invoke (hidden from / menu)
allowed-tools: Read, Grep, Glob # Restrict tool access
model: opus # Model override (sonnet, opus, haiku)
context: fork # Run in isolated subagent
agent: Explore # Subagent type when context: fork
hooks: ... # Skill-scoped hooks
---All fields optional. Only description is recommended.
String Substitutions
Formatting purposes: all times it would be ! I have written ❗️.
| Variable | Description |
|---|---|
$ARGUMENTS | All arguments passed to skill |
$ARGUMENTS[N] / $N | Specific argument (0-indexed) |
${CLAUDE_SESSION_ID} | Current session ID |
` ❗️command ` | Dynamic injection - runs before sending |
If $ARGUMENTS not in content, args appended as ARGUMENTS: <value>.
Invocation Control
| Setting | User invokes | Claude invokes | In context |
|---|---|---|---|
| (default) | Yes | Yes | Description |
disable-model-invocation: true | Yes | No | Not loaded |
user-invocable: false | No | Yes | Description |
Subagent Execution
Add context: fork to run in isolation. Skill content becomes the subagent's task.
---
name: deep-research
context: fork
agent: Explore # or Plan, general-purpose, or custom agent name
---
Research $ARGUMENTS thoroughly...Built-in agents: Explore, Plan, general-purpose. Custom agents from .claude/agents/.
Dynamic Context Injection
Shell commands run BEFORE content sent to Claude:
---
name: pr-summary
---
PR diff: ❗️`gh pr diff`
PR comments: ❗️`gh pr view --comments`
Summarize this PR...Extended Thinking
Include "ultrathink" anywhere in skill content to enable extended thinking.
Common Patterns
Reference skill (conventions/guidelines):
---
name: api-conventions
description: API design patterns for this codebase
---
When writing API endpoints:
- Use RESTful naming
- Return consistent error formatsTask skill (user-triggered action):
---
name: deploy
description: Deploy to production
disable-model-invocation: true
context: fork
---
Deploy $ARGUMENTS:
1. Run tests
2. Build
3. Push to deployment targetRead-only skill (restricted tools):
---
name: safe-reader
description: Explore without modifying
allowed-tools: Read, Grep, Glob
---Permissions
Restrict in /permissions:
Skill- deny all skillsSkill(name)- exact matchSkill(name *)- prefix match with args
Troubleshooting
- Skill not triggering: Check description keywords, verify with "What skills are available?"
- Triggers too often: Make description more specific or add
disable-model-invocation: true - Missing skills: Budget is 2% of context window. Set
SLASH_COMMAND_TOOL_CHAR_BUDGETto override.
Related skills
Find SkillsSo their agent can intelligently recommend and install the right modular skills from the open ecosystem when they ask how to do something.3M27.9k
Lark Skill MakerRapidly create custom skills for Lark (Feishu) that can be invoked by their AI agents and automation workflows.513k
Skill CreatorGenerate, test, and iterate on new reusable agent skills directly inside their Claude or Cursor workflow.337k166k
BrainstormingTurn vague product ideas into clear, approved designs and specifications through guided collaborative dialogue before writing any code.307k265k
Writing Great SkillsCreate high-quality, reusable agent skills that other developers can discover and install from the Skillselion catalog.290k200k
Find SkillsInstantly surface the best relevant agent skills, MCP servers, and workflows across multiple registries before starting any new project.226k12FAQ
Where do skills live?
Personal at ~/.claude/skills, project at .claude/skills, plus plugin and enterprise locations, with priority enterprise > personal > project.
How do you control who can invoke a skill?
Frontmatter flags like disable-model-invocation and user-invocable control whether the user, Claude, or both can invoke it.