Answers from the documentation; checking which model is on duty
Hi - I answer from the OpenSmartRoute documentation: routing, the API, plans and quotas, self-hosting. Ask away, or open a support ticket if you need a person.
Grounded in the docs - follow a source before acting on it.
j-config-audit - Prompt - OpenSmartRoute
Promptv1.0.0
j-config-audit
Security audit of Claude, Codex, Gemini, and shared agent configuration for secrets, over-broad permissions, and prompt-injection vectors.
Prompt file imported from jlaws/dotfiles (.codex/prompts/j-config-audit.md). Fill in {{arguments}} before use. Copyright stays with the author.
Load the config-security-audit skill by reading ~/.agents/skills/config-security-audit/SKILL.md before doing anything else. Run the full config-security scan.
Scope: {{arguments}}
If no argument provided, scan all configuration trees.
Use it
Copy one of these into your project. Installing also returns the manifest and these snippets.
Load the `config-security-audit` skill by reading `~/.agents/skills/config-security-audit/SKILL.md` before doing anything else. Run the full config-security scan.
Scope: {{arguments}}
If no argument provided, scan all configuration trees.
Manifest
The prompt text and its fill-in variables. Copy it or fetch it by URL from your own code.
{
"prompt": "Load the `config-security-audit` skill by reading `~/.agents/skills/config-security-audit/SKILL.md` before doing anything else. Run the full config-security scan.\n\nScope: {{arguments}}\n\nIf no argument provided, scan all configuration trees.",
"variables": [
"arguments"
],
"notes": "Security audit of Claude, Codex, Gemini, and shared agent configuration for secrets, over-broad permissions, and prompt-injection vectors. Arguments: <scope: secrets|permissions|injection|path>.",
"metadata": {
"source": {
"provider": "github-codex-prompts",
"repository": "https://github.com/jlaws/dotfiles",
"path": ".codex/prompts/j-config-audit.md",
"ref": "db3261063fc7003304cd3a4666e5e786d7dde1f1",
"url": "https://github.com/jlaws/dotfiles/blob/db3261063fc7003304cd3a4666e5e786d7dde1f1/.codex/prompts/j-config-audit.md",
"key": "jlaws/dotfiles/.codex/prompts/j-config-audit.md"
}
}
}
Installed into a catalogue, chosen by a router
Install j-config-audit and it becomes one more candidate the router can pick - when it fits.
A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.