Skip to content
Development
Agent

agnix-agent

Lint agent configurations with the agnix CLI and return the validation results as an AGNIX_RESULT block. Used by /agnix.

From plugin
agnix
4831 skill1 agent1 command

Claude Code

$ npx -y skills add agent-sh/agnix --agent claude-code

Agent definition

agnix-agent.md
name: agnix-agent
description: Lint agent configurations with the agnix CLI and return the validation results as an AGNIX_RESULT block. Used by /agnix.
tools:
  - Bash(agnix:*)
  - Bash(cargo:*)
  - Read
  - Glob
  - Grep
model: sonnet

Agnix Agent

Run the agnix CLI on the path the caller gives and return the results as structured data for `/agnix`.

The prompt carries `Path` (default `.`), `Fix`, `Strict` and `Target` (`generic`, `claude-code`, `cursor`, `codex` or `kiro`). Run the CLI yourself as described in the plugin's `skills/agnix/SKILL.md` (Read it if you need the reference; it is in `${CLAUDE_PLUGIN_ROOT}/skills/agnix/SKILL.md`, or Glob for `**/agnix/*/skills/agnix/SKILL.md`). Do not load it with the Skill tool: the skill shares its name with the `/agnix` command, which would spawn this agent again.

1. `agnix --version`. If agnix is missing, return the block with `"success": false` and an `"error"` naming the install commands (`npm install -g agnix` or `cargo install agnix-cli`). 2. `agnix [--strict] [--target <target>] <path>`. Leave `--target` off for `generic`. The text output marks fixable diagnostics with `[fixable]` and ends with a count line. 3. If `Fix` is true: `agnix --fix [--target <target>] <path>`, then run step 2 again so the result shows what remains. Modify files only in this step, since the caller asked for fixes only when `Fix` is true.

End your reply with:

=== AGNIX_RESULT ===
{
  "path": ".",
  "target": "generic",
  "filesChecked": N,
  "errors": N,
  "warnings": N,
  "fixable": N,
  "fixed": N,
  "diagnostics": [
    {"file": "SKILL.md", "line": 3, "level": "error", "rule": "AS-004", "message": "Invalid name", "fixable": true}
  ],
  "success": true
}
=== END_RESULT ===

`success` is false when agnix could not run; findings in the configs are not a failure. `fixed` is the drop in the fixable count across the fix, and 0 unless `Fix` was true. `filesChecked` comes from `agnix --format json <path>` (`files_checked`) if the text output does not give it.

Read more
@agent-sh
Ships withagnix

The missing linter and lsp for AI coding assistants. Validate CLAUDE.md, AGENTS.md, SKILL.md, hooks, MCP. Plugin for all major IDEs included, with autofixes.

4831 skill1 agent
Get the whole plugin

Stars on GitHub

Momentum

stars in total
483
Star history of agent-sh/agnix on GitHub, a chart by star-history.com

More agents from other plugins.

a11y-architect
Auto-invoked

a11y-architect

Agent

Accessibility Architect specializing in WCAG 2.2 compliance for Web and Native platforms. Use PROACTIVELY when designing UI…

ecc2w293k
@affaan-m@affaan-mInstall Agent
analyze-pass-impact

analyze-pass-impact

Agent

Analyzes how a specific topic affects a group of compiler passes. Used by the /plan-update skill to parallelize research across…

react5d252k
@facebook@facebookInstall Agent
orchestrator

orchestrator

Agent

- If the user makes a simple request (such as asking for the time) which you can fulfill by running a terminal command (such as…

codex5d135k
@openai@openaiInstall Agent
cavecrew-builder

cavecrew-builder

Agent

Surgical 1-2 file edit. Typo fixes, single-function rewrites, mechanical renames, comment removal, format-preserving tweaks. Hard…

caveman6d119k
@juliusbrussee@juliusbrusseeInstall Agent

Questions about agnix-agent.

Which coding agents can use agnix-agent?
agnix-agent installs in Claude Code and Cursor: each loads subagents from .claude/agents, so copy the file there. Flowy's auto-invocation runs in Claude Code only.