Agent Setup Health Audit
tw93/Waza
Audits a project's agent configuration, instruction drift, hooks, MCP and AI maintainability, then reports prioritized findings with evidence and next actions.
How to answer questions about the agentica product you are running inside — CLI flags, config.yaml profiles, API keys, models, sessions, resume, workspace, AGENTS.md standing rules, skills, logs…
Install Agentica
$ npx skills add shibing624/agentica --skill agentica -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install shibing624/agentica agentica --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/shibing624/agentica.git skills-src && mkdir -p .claude/skills && cp -r skills-src/agentica/skills/bundled/agentica .claude/skills/agentica && 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 "agentica" agent skill from https://github.com/shibing624/agentica/tree/main/agentica/skills/bundled/agentica into .claude/skills/agentica/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentica", 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/shibing624/agentica/tree/main/agentica/skills/bundled/agenticaType 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 shibing624/agentica --skill agentica -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install shibing624/agentica agentica --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/shibing624/agentica.git skills-src && mkdir -p .agents/skills && cp -r skills-src/agentica/skills/bundled/agentica .agents/skills/agentica && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "agentica" agent skill from https://github.com/shibing624/agentica/tree/main/agentica/skills/bundled/agentica into .agents/skills/agentica/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentica", 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 shibing624/agentica --skill agentica -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install shibing624/agentica agentica --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/shibing624/agentica.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/agentica/skills/bundled/agentica .cursor/skills/agentica && 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 "agentica" agent skill from https://github.com/shibing624/agentica/tree/main/agentica/skills/bundled/agentica into .cursor/skills/agentica/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentica", 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/shibing624/agentica.git --path agentica/skills/bundled/agentica--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 shibing624/agentica --skill agentica -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install shibing624/agentica agentica --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/shibing624/agentica.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/agentica/skills/bundled/agentica .gemini/skills/agentica && 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 "agentica" agent skill from https://github.com/shibing624/agentica/tree/main/agentica/skills/bundled/agentica into .gemini/skills/agentica/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentica", 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 shibing624/agentica agenticaInstalls 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 shibing624/agentica --skill agentica -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/shibing624/agentica.git skills-src && mkdir -p .github/skills && cp -r skills-src/agentica/skills/bundled/agentica .github/skills/agentica && 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 "agentica" agent skill from https://github.com/shibing624/agentica/tree/main/agentica/skills/bundled/agentica into .github/skills/agentica/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentica", 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 shibing624/agentica --skill agentica -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install shibing624/agentica agentica --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/shibing624/agentica.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/agentica/skills/bundled/agentica .opencode/skills/agentica && 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 "agentica" agent skill from https://github.com/shibing624/agentica/tree/main/agentica/skills/bundled/agentica into .opencode/skills/agentica/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentica", 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
agentica- GitHub stars
- 352
- Token cost
- ~1.8k tokens
- SKILL.md length
- 1,003 words
- Files
- 1
- Skills in repo
- 4
- Repo updated
- First seen
- Licence
- Apache-2.0
At a glance
How to answer questions about the agentica product you are running inside — CLI flags, config.yaml profiles, API keys, models, sessions, resume, workspace, AGENTS.md standing rules, skills, logs…
- Asked how agentica works
- SKILL.md covers Rule 1: look it up, do not…, Rule 2: you cannot type slash…, The map and Config concepts that do not…, plus 3 more sections
- Calls pip
- How to configure
What it does
Agentica is an agent skill from shibing624/agentica. How to answer questions about the agentica product you are running inside — CLI flags, config.yaml profiles, API keys, models, sessions, resume, workspace, AGENTS.md standing rules, skills, logs, upgrade, and selfmanage. Use when asked how agentica works, how to configure or upgrade it, where state lives on disk, where to write "remember: always X", or how to change your own settings.
Its SKILL.md is about 1.8k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Agent Workflows, covering Agent instruction files and Hooks and plugins. The repository describes itself as: One person, a team of agents. Multi-session CLI that collaborates across terminals; /goal keeps long tasks running; WeChat/WeCom/Feishu gateway lets you call them back when you… The licence is Apache-2.0.
When your agent uses it
- Asked how agentica works
- How to configure
- Where state lives on disk
- Where to write remember: always X
Example prompts
- “remember: always X”
- “/agentica”
Requirements
- Python 3
What it can do on your machine
Read from SKILL.md and the folder at commit 0b5e227. 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:
pip
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use pip, 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
Agentica loads about 1.8k tokens when it runs. Until then it costs about 99 tokens; SKILL.md has 1,003 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: notes
The automated check noted patterns worth knowing about, such as sudo or a known installer.
- NoteMentions a .env fileSKILL.md:53
| `.env` | hand-maintained keys, loaded in addition to `config.yaml` | - NoteMentions a .env fileSKILL.md:68
**Precedence.** Shell environment beats `.env`, which beats `config.yaml`. A - NoteMentions a .env fileSKILL.md:81
and `.env`, reports and installs upgrades, and installs skills. Its own schema
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 shibing624/agentica at commit 0b5e227, republished under its Apache-2.0 licence (© shibing624). 1,003 words, ~1,781 tokens.
Download SKILL.mdSave it as
.claude/skills/agentica/SKILL.md (or your agent's skills folder).- name
- agentica
- description
- How to answer questions about the agentica product you are running inside — CLI flags, config.yaml profiles, API keys, models, sessions, resume, workspace, AGENTS.md standing rules, skills, logs, upgrade, and self_manage. Use when asked how agentica works, how to configure or upgrade it, where state lives on disk, where to write "remember: always X", or how to change your own settings.
- metadata.version
- 1.0
Driving agentica
You are almost certainly running inside an agentica CLI session right now. This skill tells you where to look things up, not what the answers are: flags, commands and tool names change with every release, so anything written down here would start lying within weeks.
Rule 1: look it up, do not recall
Three live sources, in the order you should reach for them:
| Question | Source | How you read it |
|---|---|---|
| What can the command line do? | agentica --help | execute |
| How is this machine configured? | your own config | self_manage(action="show") |
| What is on disk, exactly? | ~/.agentica/config.yaml | read_file |
| What is this session doing? | slash commands | ask the user to type it |
agentica --help is generated from the parser, so it is never out of date. It
is the authoritative answer for every flag, every subcommand and every value an
option accepts. Read it before answering any "can agentica ...?" question.
Rule 2: you cannot type slash commands
/help, /status, /model, /tools, /skills, /resume, /permissions and
the rest are handled by the CLI's input loop, before anything reaches you. A
slash command in your reply is just text, and a slash command inside a peer
message is just text too.
When the answer lives behind one, say which one the user should type. /help
lists all of them and is the source of truth for what exists.
The map
Everything agentica keeps is under ~/.agentica (or $AGENTICA_HOME):
| Path | What it holds |
|---|---|
config.yaml | named profiles + active_profile; the file is hand-editable and comments survive writes |
.env | hand-maintained keys, loaded in addition to config.yaml |
logs/ | one log per CLI process; often the fastest way to see what another session actually did |
workspace/ | long-term memory; standing rules in users/<id>/AGENTS.md (see below) |
skills/ | user-installed skills, one directory per skill with a SKILL.md |
projects/ | session transcripts as JSONL, partitioned per working directory |
cache/peers/ | the live-session directory and mailboxes behind list_agents / send_message |
Read these files directly when you need a fact about the current setup. Never print an API key you read from them, and never write a key into a command line.
Config concepts that do not change
- Profiles.
config.yamlholds several named profiles;active_profilepicks one. A profile carries provider, model, base URL, key, and optional tuning. Switching profiles switches all of it at once. - Precedence. Shell environment beats
.env, which beatsconfig.yaml. A variable already set in the environment is never overwritten. - Main and auxiliary model. The main model answers the user. The auxiliary
model is the cheap one used for background work nobody is waiting on: memory
extraction, context compression, and the
tasksubagent. A profile that omits the auxiliary block reuses the main model for both. - Permission tiers. Tool access runs in one of three modes. The user can change it mid-session, so never assume the tier you started with is still in force; if a tool is refused, that is the answer, not a bug to work around.
Changing yourself
Reading is files; changing is the self_manage tool. It edits config.yaml
and .env, reports and installs upgrades, and installs skills. Its own schema
lists the actions and is in front of you every turn, so it — not this page — is
where you look up what to pass.
- Do not hand-edit
config.yamlor invent apipcommand. Writing the file yourself loses the comment-preserving round trip; runningpip install -Uthroughexecuteskips the version check and the restart notice. - Keys never touch a command line, a log, or your reply.
set_configandset_envtake them as arguments;showmasks them on the way back. upgradeneedsconfirm=Trueand installs a new version. Say what changes before asking for that confirmation, and afterwards tell the user to restart the CLI — the running process keeps the old code.- Config edits land in the file immediately; the live session does not change. A new model or tuning value applies on the next agent rebuild or restart, so say so rather than implying the switch already happened.
/configand/upgradeare the human's version of this tool. Point the user at them when they want to drive; use the tool when you are the one doing it.
Show full SKILL.md (315 more words)Show less
Standing rules: AGENTS.md (not self_manage)
When the user says "remember: always X" / "from now on ..." / "never ...", that
is a standing instruction, not a config.yaml field and not a save_memory
fact. There is no dedicated tool: append a line with apply_patch (or
write_file if the file is missing).
| Scope | File | Who sees it |
|---|---|---|
| This user, every project | ~/.agentica/workspace/users/<user_id>/AGENTS.md (CLI is default, so ~/.agentica/AGENTS.md is also a symlink to .../users/default/AGENTS.md) | every later session of this user |
| This repository only | <repo root>/AGENTS.md | sessions started anywhere under that repo; the user may commit it |
The path of every file already in the system prompt appears as <!-- /abs/path -->
above it — reuse that path, do not guess. There is no workspace-root
~/.agentica/workspace/AGENTS.md. User rules live under
users/<id>/AGENTS.md; for the default CLI user only, ~/.agentica/AGENTS.md
is a compatibility symlink to that same file.
This session: the user's request (and your write) are already in the conversation history, so follow the rule for the rest of the turn/session without waiting for the system prompt to change. Next session: the AGENTS.md chain is read once at session start into the system prompt, so the new line is there automatically.
Facts ("I am a data scientist", "the deploy target is X") still go to
save_memory. The MEMORY.md index (title + hook) is in the system prompt;
read the topic file or call search_memory for the body.
Answering well
State which source you checked. "Per agentica --help on this machine" is
worth more than a confident paragraph, because the user's installed version is
the only version that matters. If --help and your memory disagree, --help
is right and your memory is stale.
For running more than one agentica at a time, or getting two sessions to talk,
use the multi-agent skill instead. Isolating this session in a git
worktree of the current repo is the worktree skill. Scheduling unattended
jobs is the cron skill (the cronjob tool).
© shibing624, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in agentica/skills/bundled/agentica of shibing624/agentica.
Open the folder on GitHubat commit 0b5e227
Compare with similar skills
Agentica 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 |
|---|---|---|---|---|---|---|
| Agentica this skillshibing624/agentica | 352 | — | ~1.8k | Automated safety check: Notes | Apache-2.0 | |
| Agent Setup Health Audittw93/Waza | 7.2k | — | ~5.2k | Automated safety check: Notes | MIT | |
| Working With Claude Code Docsobra/superpowers-developing-for-claude-code | 142 | — | ~1.5k | Automated safety check: Pass | None | |
| Directional Promptingkingbootoshi/directional-prompting | 143 | — | ~2.3k | Automated safety check: Pass | MIT | |
| Claude Code Mastery Squadohmyjahh/xquads-squads | 277 | — | ~1.1k | Automated safety check: Pass | MIT | |
| Vault Setupearlyaidopters/second-brain | 194 | — | ~1.4k | Automated safety check: Pass | None |
Similar skills
Working With Claude Code Docs
obra/superpowers-developing-for-claude-code
Looks up official Claude Code documentation stored as reference files instead of guessing about CLI commands, configuration, or plugin APIs.
Directional Prompting
kingbootoshi/directional-prompting
Write prompts, system instructions, agent directives, slash commands, and skill descriptions using two stacked layers — outcome-first (define the destination, success criteria, stopping condition)…
Claude Code Mastery Squad
ohmyjahh/xquads-squads
Routes a request to one of eight specialist agents covering hooks, skills, subagents, MCP integration and context engineering for Claude Code.
Vault Setup
earlyaidopters/second-brain
Interactive Obsidian vault configurator. An agent skill from earlyaidopters/second-brain.
- Official
Mistral Vibe CLI Reference
mistralai/mistral-vibe
Reference for Mistral Vibe, the CLI agent it runs inside: config files, env vars, agents, skills, tools, hooks and MCP servers, so the agent can explain and troubleshoot its own setup.
More from shibing624/agentica
Multi Agent
shibing624/agentica
Run work across more than one agent — choose between the task subagent, the delegate tool, and a second agentica CLI in tmux driven by peer messages (listagents / sendmessage).
Worktree
shibing624/agentica
Isolate parallel work in a git worktree of the current repository (worktree tool: status / new / merge / remove).
Cron
shibing624/agentica
Schedule, list, pause, resume, edit, or run the user's agentica cron jobs with the cronjob tool.
Related
Categories
Questions about Agentica
What does Agentica do?
How to answer questions about the agentica product you are running inside — CLI flags, config.yaml profiles, API keys, models, sessions, resume, workspace, AGENTS.md standing rules, skills, logs…. Agentica is an agent skill from shibing624/agentica.md standing rules, skills, logs, upgrade, and selfmanage.
When should I use Agentica?
Agentica fits situations like: asked how agentica works; how to configure; where state lives on disk; where to write remember: always X.
How do I install Agentica in Claude Code?
Run `npx skills add shibing624/agentica --skill agentica -a claude-code`. Or copy the skill folder (agentica/skills/bundled/agentica in shibing624/agentica) into .claude/skills/agentica in your project. Claude Code loads it when a task matches its description.
How do I install Agentica in Codex?
Run `npx skills add shibing624/agentica --skill agentica -a codex`. Or copy the skill folder (agentica/skills/bundled/agentica in shibing624/agentica) into .agents/skills/agentica in your project. Codex loads it when a task matches its description.
Can I use Agentica 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 shibing624/agentica --skill agentica -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/agentica, .gemini/skills/agentica, .github/skills/agentica and .opencode/skills/agentica in your project.
What does Agentica need to run?
Going by SKILL.md and its folder, Agentica needs the command-line tools its instructions call (pip). Our summary lists: Python 3.
Does Agentica access the network?
SKILL.md contains no URLs. Its commands use pip, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Is Agentica safe to install?
Our automated static check of SKILL.md found notes only (mentions a .env file), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.
What licence does Agentica use?
Agentica is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Agentica use?
About 1.8k tokens (SKILL.md is roughly 7.1k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Agentica?
Skills that share tags, products or a category with Agentica: Agent Setup Health Audit (tw93/Waza, 7.2k stars), Working With Claude Code Docs (obra/superpowers-developing-for-claude-code, 142 stars), Directional Prompting (kingbootoshi/directional-prompting, 143 stars) and Claude Code Mastery Squad (ohmyjahh/xquads-squads, 277 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Agentica?
shibing624 (a GitHub user) maintains it in shibing624/agentica, which has 352 GitHub stars. The repository holds 4 skills in this directory. The repository was last updated on October 9, 2026.
Source: shibing624/agentica on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.