# Manage Skills

> Create, update, inspect, or manage reusable profile skills with skill_manage. Use when the user wants the agent to remember a repeatable workflow, change a skill, or maintain skill instructions.

- **Type:** Skill
- **Install:** `agentstack add skill-ahmadrosid-nakama-manage-skills`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [ahmadrosid](https://agentstack.voostack.com/s/ahmadrosid)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [ahmadrosid](https://github.com/ahmadrosid)
- **Source:** https://github.com/ahmadrosid/nakama/tree/main/packages/core/src/skills/bundled/manage-skills
- **Website:** https://ahmadrosid.github.io/nakama

## Install

```sh
agentstack add skill-ahmadrosid-nakama-manage-skills
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

Use skills for repeatable procedures and workflows the agent should execute later. Do not use skills for user facts, preferences, or observations; use the `update-profile-memory` skill for those.

## Primary path — skill_manage

When the `skill_manage` tool is available, use it for all skill create/update/delete work and for supporting files under a skill directory:

| Action | When |
|--------|------|
| `install` | Import public GitHub SKILL.md — pass `url`; see `skill-installer` for discovery and limitations |
| `create` | New reusable workflow — pass full SKILL.md (`content`) |
| `patch` | Targeted fix — `name`, `old_string`, `new_string` (preferred over rewrite) |
| `edit` | Full SKILL.md replacement — `name` + `content` (frontmatter name must match) |
| `delete` | Remove a profile-owned skill — `name` |
| `write_file` | Supporting file under the skill dir — `name`, `path`, `content` |
| `remove_file` | Remove a supporting file — `name`, `path` |

`skill_manage` writes under the profile skills directory, validates frontmatter, and **auto-assigns** the skill so it can match on later turns. Bundled and global skills are read-only. `write_file` / `remove_file` refuse `SKILL.md` and skill-local tools (`tool.ts` / `tool.js`).

When **write approval** is enabled for the org or profile, `skill_manage` **stages** changes instead of writing immediately. The tool returns `{ staged: true, proposalId, … }` and an org admin must approve the proposal before the skill goes live. Do not re-submit identical pending proposals; wait for review or ask an admin.

Example create content:

```markdown
---
name: skill-name
description: Short trigger description explaining when to use this skill.
include-body-on-match: true
---

Step-by-step instructions for the repeatable workflow.
```

Use lowercase kebab-case names. Prefer `patch` for small updates; use `edit` for full rewrites. Do not create broad or vague skills; avoid storing private user facts.

## File tools (secondary)

When `skill_manage` is present, `write_file` / `edit_file` / `delete_file` **cannot** write under `skills/*/…` — they refuse with a redirect to `skill_manage`. You may still use file tools to inspect skills (`read_file`, `search_files`).

If `skill_manage` is unavailable, do not invent a skill by writing orphan `SKILL.md` files that will not auto-assign.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [ahmadrosid](https://github.com/ahmadrosid)
- **Source:** [ahmadrosid/nakama](https://github.com/ahmadrosid/nakama)
- **License:** MIT
- **Homepage:** https://ahmadrosid.github.io/nakama

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-ahmadrosid-nakama-manage-skills
- Seller: https://agentstack.voostack.com/s/ahmadrosid
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.