Using Agent Skills
addyosmani/agent-skills
Meta-skill for choosing which workflow skill fits the task at hand, plus always-on habits: surface assumptions, stop on confusion, push back, keep it simple and stay in scope.
A skill your agent uses when: initializing, auditing, or optimizing repository AI guidance, including AGENTS.md/CLAUDE.md bridges, Agent Skills, trigger descriptions, progressive disclosure, source…
Install AI Agent Maintenance
$ npx skills add owent/libatbus --skill ai-agent-maintenance -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install owent/libatbus ai-agent-maintenance --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/owent/libatbus.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/ai-agent-maintenance .claude/skills/ai-agent-maintenance && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "ai-agent-maintenance" agent skill from https://github.com/owent/libatbus/tree/main/.agents/skills/ai-agent-maintenance into .claude/skills/ai-agent-maintenance/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-agent-maintenance", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/owent/libatbus/tree/main/.agents/skills/ai-agent-maintenanceType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add owent/libatbus --skill ai-agent-maintenance -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install owent/libatbus ai-agent-maintenance --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/owent/libatbus.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/ai-agent-maintenance .agents/skills/ai-agent-maintenance && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "ai-agent-maintenance" agent skill from https://github.com/owent/libatbus/tree/main/.agents/skills/ai-agent-maintenance into .agents/skills/ai-agent-maintenance/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-agent-maintenance", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add owent/libatbus --skill ai-agent-maintenance -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install owent/libatbus ai-agent-maintenance --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/owent/libatbus.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/ai-agent-maintenance .cursor/skills/ai-agent-maintenance && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "ai-agent-maintenance" agent skill from https://github.com/owent/libatbus/tree/main/.agents/skills/ai-agent-maintenance into .cursor/skills/ai-agent-maintenance/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-agent-maintenance", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/owent/libatbus.git --path .agents/skills/ai-agent-maintenance--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add owent/libatbus --skill ai-agent-maintenance -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install owent/libatbus ai-agent-maintenance --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/owent/libatbus.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/ai-agent-maintenance .gemini/skills/ai-agent-maintenance && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "ai-agent-maintenance" agent skill from https://github.com/owent/libatbus/tree/main/.agents/skills/ai-agent-maintenance into .gemini/skills/ai-agent-maintenance/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-agent-maintenance", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install owent/libatbus ai-agent-maintenanceInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add owent/libatbus --skill ai-agent-maintenance -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/owent/libatbus.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/ai-agent-maintenance .github/skills/ai-agent-maintenance && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "ai-agent-maintenance" agent skill from https://github.com/owent/libatbus/tree/main/.agents/skills/ai-agent-maintenance into .github/skills/ai-agent-maintenance/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-agent-maintenance", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add owent/libatbus --skill ai-agent-maintenance -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install owent/libatbus ai-agent-maintenance --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/owent/libatbus.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/ai-agent-maintenance .opencode/skills/ai-agent-maintenance && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "ai-agent-maintenance" agent skill from https://github.com/owent/libatbus/tree/main/.agents/skills/ai-agent-maintenance into .opencode/skills/ai-agent-maintenance/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-agent-maintenance", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
ai-agent-maintenance- GitHub stars
- 235
- Token cost
- ~1.4k tokens
- SKILL.md length
- 677 words
- Files
- 2 (incl. references)
- Skills in repo
- 6
- Repo updated
- First seen
- Licence
- MIT
At a glance
A skill your agent uses when: initializing, auditing, or optimizing repository AI guidance, including AGENTS.md/CLAUDE.md bridges, Agent Skills, trigger descriptions, progressive disclosure, source…
- Works in 5 steps: Discover before loading → Verify facts and choose process strength → Choose one authoritative surface → …
- Optimizing repository AI guidance
- SKILL.md covers Outcome, Workflow, Repository invariants and References
- Calls git
What it does
AI Agent Maintenance is an agent skill from owent/libatbus. Use when: initializing, auditing, or optimizing repository AI guidance, including AGENTS.md/CLAUDE.md bridges, Agent Skills, trigger descriptions, progressive disclosure, source tracking, or cross-tool compatibility. Do not use when: only implementing code under existing guidance.
Its SKILL.md is about 1.4k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files, including reference files (for example `references/compatibility-sources.md`).
It sits in Agent Workflows, covering Agent instruction files. The repository describes itself as: 用于搭建高性能、全异步、树形结构的BUS消息系统的跨平台框架库. The licence is MIT.
When your agent uses it
- Optimizing repository AI guidance
- Including AGENTS.md/CLAUDE.md bridges
- Trigger descriptions
- Progressive disclosure
Example prompts
- “/ai-agent-maintenance”
Workflow steps
5 steps, taken from the step headings in SKILL.md.
What it can do on your machine
Read from SKILL.md and the folder at commit d46c04d. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
git
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
AI Agent Maintenance loads about 1.4k tokens when it runs, and up to ~2.1k if it reads all its reference files. Until then it costs about 76 tokens; SKILL.md has 677 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from owent/libatbus at commit d46c04d, republished under its MIT licence (© owent). 677 words, ~1,448 tokens.
Download SKILL.mdSave it as
.claude/skills/ai-agent-maintenance/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.- name
- ai-agent-maintenance
- description
- Use when: initializing, auditing, or optimizing repository AI guidance, including AGENTS.md/CLAUDE.md bridges, Agent Skills, trigger descriptions, progressive disclosure, source tracking, or cross-tool compatibility. Do not use when: only implementing code under existing guidance.
AI Agent Maintenance
Use this Skill for AI-facing surfaces in libatbus. Keep this repository self-contained; do not depend on parent,
sibling, or vendored-submodule prompt files.
Outcome
- Keep AI-facing guidance source-backed, compact, task-routed, and independently usable in this repository.
- Merge improvements into the current authoritative files; remove stale, duplicated, or historical variants.
- Preserve the user's dirty workspace and avoid tool-specific configuration without a verified need.
Workflow
1. Discover before loading
- Inspect the active harness's actual Skills, tools, modes, permissions, and extensions. Do not assume another client's capabilities or install a workflow without user authorization.
- Read
AGENTS.md,CLAUDE.md,.agents/skills/README.md, repository status, and Skillname/descriptionmetadata first. Inspect whether tool-specific directories or AI docs exist; do not preload every Skill body or reference. - Load a full
SKILL.md, bundled resource, legacy prompt, or client-specific config only when the task or index routes there. Read generated/template inputs when they are the real source of truth.
2. Verify facts and choose process strength
- Separate stable repository facts from mutable external behavior. Verify the former in current code/config/tests and the latter in current official documentation before writing compatibility or security claims.
- Treat OpenSpec and Superpowers as optional, mutable integrations. Verify repository adoption and the active harness's current capabilities before referencing their artifacts or commands; never install or initialize them implicitly.
- State assumptions, the smallest sufficient plan, rollback, and validation after the source pass.
- Use the shortest verified path for small, low-risk edits. Require reviewable scope and acceptance criteria before cross-module behavior, public API, data model, security, or deployment changes; reuse existing project artifacts rather than initializing a methodology for ceremony.
3. Choose one authoritative surface
- Put stable, high-signal, broadly applicable rules in
AGENTS.md. - Keep
CLAUDE.mdas a thin@AGENTS.mdand Skill-index bridge; add only verified Claude-specific differences. - Keep
.agents/skills/README.mdto names, one-line routing, and compact maintenance rules. - Put repeatable, specialized procedures and non-obvious project knowledge in
.agents/skills/<name>/SKILL.md. - Add client-specific rules, agents, prompts, or Skill mirrors only when the shared surfaces cannot express a verified requirement. Write only the difference and do not duplicate shared rules.
Show full SKILL.md (339 more words)Show less
4. Maintain Skills for low context cost
- Keep folder and
nameidentical, lowercase, hyphenated, and at most 64 characters. Keep portable frontmatter tonameanddescriptionunless a verified client-specific requirement justifies a separate client-owned surface. - Make
descriptionintent-first and specific: say what the Skill does, when it should trigger, and a useful near-miss boundary. Keep the key trigger early and stay within 1024 characters. - Write imperative, project-specific procedures. Remove explanations the agent already knows; use one clear default and make control stricter only where the workflow is fragile.
- Keep the main body under 500 lines and about 5,000 tokens. Move detailed APIs, examples, and variants to one-level
references/; keep deterministic repeated logic in tested, non-interactivescripts/with actionable errors. - For a new or materially changed trigger, draft 8-10 realistic should-trigger queries and 8-10 near-miss should-not-trigger queries. Measure invocation only when the active client exposes observable Skill calls; otherwise perform a manual boundary review and report that no trigger rate was measured.
5. Validate and report
- Run an available Agent Skills validator; otherwise check YAML delimiters, required fields, folder/name equality, description length, referenced paths, and absence of placeholders directly.
- Run markdown diagnostics, line-ending/trailing-whitespace checks, and scoped
git diff --check. Re-read routing files to confirm bridges remain thin and references load only on demand. - Validate from every affected nested Git root. Run code/build/tests only when behavior or executable content changed; for documentation-only work, state what was skipped and why.
- Report changed surfaces, verified sources, capability boundaries, validation results, and unresolved risks.
Repository invariants
- Keep engineering conventions in
../engineering-guidelines/SKILL.md; route to them instead of duplicating details. - Resolve
<BUILD_DIR>from the nearest VS Code CMake setting, clangd compile-command path, or existing configured tree; fall back tobuild. Put scratch and script/log output under<BUILD_DIR>/_agent_tmp/.... - Keep this repository self-contained; do not mirror rules into parent, sibling, or vendored-submodule prompts.
References
- Read compatibility sources only when changing Skill formats, discovery paths, bridge behavior, client-specific configuration, MCP guidance, or cross-tool compatibility claims.
- Read upstream change-workflow methods only when changing risk routing or OpenSpec/Superpowers integration policy.
© owent, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 1 other file (references) in .agents/skills/ai-agent-maintenance of owent/libatbus.
- SKILL.md
- references/compatibility-sources.md
Open the folder on GitHubat commit d46c04d
Compare with similar skills
AI Agent Maintenance next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| AI Agent Maintenance this skillowent/libatbus | 235 | — | ~1.4k | Automated safety check: Pass | MIT | |
| Using Agent Skillsaddyosmani/agent-skills | 103k | 4 repos | ~2.4k | Automated safety check: Pass | MIT | |
| Claude ReflectBayramAnnakov/claude-reflect | 1.7k | 2 repos | ~627 | Automated safety check: Pass | MIT | |
| Neat-Freak Knowledge CloseoutKKKKhazix/khazix-skills | 21k | — | ~1.9k | Automated safety check: Pass | MIT | |
| Writing For Agentsbestofjs/bestofjs | 3.1k | 18 repos | ~2.7k | Automated safety check: Pass | MIT | |
| Task Observerrebelytics/one-skill-to-rule-them-all | 3.2k | 1 repos | ~12k | Automated safety check: Pass | CC-BY-4.0 |
Similar skills
Claude Reflect
BayramAnnakov/claude-reflect
Self-learning system that captures corrections during sessions and reminds users to run /reflect to update CLAUDE.md.
Neat-Freak Knowledge Closeout
KKKKhazix/khazix-skills
Brings project docs, agent rule files, authorized memory and leftover workspace files back in line with what the code and runtime actually do at the end of a work session.
Writing For Agents
bestofjs/bestofjs
Writing documents for agents. An agent skill from bestofjs/bestofjs.
Task Observer
rebelytics/one-skill-to-rule-them-all
Monitors task execution for skill improvement opportunities.
- Official
SkillOpt Sleep Cycle
microsoft/SkillOpt
Runs an on-demand or nightly sleep cycle that reviews past Claude Code sessions and proposes validated updates to CLAUDE.md and skills.
More from owent/libatbus
Testing
owent/libatbus
A skill your agent uses when: designing, writing, reviewing, or running libatbus unit tests, filtering private-framework cases, testing topology/channel behavior, or diagnosing Windows test…
Libatbus Protocol Crypto
owent/libatbus
A skill your agent uses when: working on libatbus protocol transport, ECDH handshakes, cipher/compression negotiation, message framing, access token auth, connectioncontext, or crypto-related tests.
Change Workflow
owent/libatbus
A skill your agent uses when: diagnosing or fixing a defect, test/build/runtime failure, or planning and implementing a nontrivial feature, behavior change, cross-module refactor, public API/ABI…
Shell Tooling
owent/libatbus
A skill your agent uses when: running terminal commands, writing or debugging shell/PowerShell scripts, choosing or installing CLI tools, or diagnosing command failures, quoting/escaping errors, or…
Engineering Guidelines
owent/libatbus
A skill your agent uses when: writing or reviewing C++, public headers, template function visibility, exported API ABI, ATFWUTILSYMBOLVISIBLE, or ATFWUTILFORCEINLINE.
Related
Categories
Questions about AI Agent Maintenance
What does AI Agent Maintenance do?
A skill your agent uses when: initializing, auditing, or optimizing repository AI guidance, including AGENTS.md/CLAUDE.md bridges, Agent Skills, trigger descriptions, progressive disclosure, source…. AI Agent Maintenance is an agent skill from owent/libatbus.md bridges, Agent Skills, trigger descriptions, progressive disclosure, source tracking, or cross-tool compatibility.
When should I use AI Agent Maintenance?
AI Agent Maintenance fits situations like: optimizing repository AI guidance; including AGENTS.md/CLAUDE.md bridges; trigger descriptions; progressive disclosure.
How do I install AI Agent Maintenance in Claude Code?
Run `npx skills add owent/libatbus --skill ai-agent-maintenance -a claude-code`. Or copy the skill folder (.agents/skills/ai-agent-maintenance in owent/libatbus) into .claude/skills/ai-agent-maintenance in your project. Claude Code loads it when a task matches its description.
How do I install AI Agent Maintenance in Codex?
Run `npx skills add owent/libatbus --skill ai-agent-maintenance -a codex`. Or copy the skill folder (.agents/skills/ai-agent-maintenance in owent/libatbus) into .agents/skills/ai-agent-maintenance in your project. Codex loads it when a task matches its description.
Can I use AI Agent Maintenance in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add owent/libatbus --skill ai-agent-maintenance -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/ai-agent-maintenance, .gemini/skills/ai-agent-maintenance, .github/skills/ai-agent-maintenance and .opencode/skills/ai-agent-maintenance in your project.
What does AI Agent Maintenance need to run?
Going by SKILL.md and its folder, AI Agent Maintenance needs the command-line tools its instructions call (git).
Does AI Agent Maintenance access the network?
SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Is AI Agent Maintenance safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does AI Agent Maintenance use?
AI Agent Maintenance is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does AI Agent Maintenance use?
About 1.4k tokens (SKILL.md is roughly 5.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 648 tokens, read only when the agent opens those files.
What are the alternatives to AI Agent Maintenance?
Skills that share tags, products or a category with AI Agent Maintenance: Using Agent Skills (addyosmani/agent-skills, 103k stars), Claude Reflect (BayramAnnakov/claude-reflect, 1.7k stars), Neat-Freak Knowledge Closeout (KKKKhazix/khazix-skills, 21k stars) and Writing For Agents (bestofjs/bestofjs, 3.1k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains AI Agent Maintenance?
owent (a GitHub user) maintains it in owent/libatbus, which has 235 GitHub stars. The repository holds 6 skills in this directory. The repository was last updated on September 11, 2026.
Source: owent/libatbus on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.