For Agents

Give coding agents a clean way to audit spend.

Xerg exposes one canonical skill, one hosted MCP server, and a small public wire contract so agent tools can run local audits, read recommendations, and query hosted workspace data.

Copy/paste to your agent and go:

Audit data stays on your machine unless you choose to push it.

See all setup paths →

Works from any terminal-capable agent.

What Xerg audits

OpenClaw, Hermes, Claude Code, Cursor usage exports, QM (admin setup), and any framework via xerg ingest event payloads. The agent you run setup from and the source Xerg audits don’t have to be the same — a Codex user audits whatever doctor finds on the machine.

Full source docs →

All agent surfaces

The setup prompt above is the recommended path. These are the interfaces Xerg exposes — the skill, hosted MCP, the CLI, and public schemas — if your tooling needs to talk to one directly.

SKILL.md

One canonical skill, distributed through the public xergai/skills repo, ClawHub @xerg/xerg, xerg.ai/skill.md, and the npm CLI bundle. skills.sh and Hermes discovery index those sources. Coordinated releases check canonical bytes; listing presence alone is not an integrity check. The skill asks before installing anything and before analyzing your data. It selects an exact reviewed CLI version and preserves that version in follow-up commands; upgrades are not automatic.

Install a reviewed skill revision →

The installation table covers terminal agents, Agent Skills hosts, OpenClaw, Hermes, and local imports. Agent Skills registration uses skills@1.5.25 and an operator-reviewed full repository SHA. Native installation paths were checked with OpenClaw 2026.9.4 on Node 24.16.0 and Hermes 0.21.0 on Python 3.12.14; see the exact source pin and review steps in the host installation guide.

Hosted MCP

Team, Growth, Scale, and Enterprise workspaces connect any remote MCP client to https://mcp.xerg.ai/mcp with a workspace API key. Run xerg mcp-setup to print or write the config for your client.

CLI

Follow the skill's pinned, non-interactive path with no account and no hosted auth. Nothing leaves the machine unless you explicitly push.

Review, installation or package download, local inspection, workspace connection, payload preview, and upload are separately approved stages.

Wire schemas

MIT-licensed @xerg/schemas on npm: TypeScript contracts for push payloads and recommendations. Build against the wire without depending on the proprietary CLI.

Installation and discovery

Host or surface	Installation or discovery path	Review boundary
Terminal-capable agent	Canonical skill	Review fetched content before use; this mutable URL does not pin a revision or install it persistently.
Agent Skills hosts	Public repository, discoverable on skills.sh	Use `skills@1.5.25` and an operator-reviewed full repository SHA; keep host and scope selection interactive.
OpenClaw	ClawHub `@xerg/xerg`	Select and review an exact skill version and its files; never substitute the bare `xerg` slug.
Hermes	Legacy discovery index or a reviewed raw GitHub Markdown URL	A full repository SHA in the raw URL fixes the source revision; the discovery index and hosted skill URL are mutable.
Local folder import	The Xerg `SKILL.md` bundled inside the approved npm CLI package	Review the bundled file independently; its instructions can differ from the current skill release.

Installation, package download, local inspection, remote access, workspace connection, payload preview, and upload require their own applicable approvals. A listing or successful installation does not certify the skill, authorize execution, or establish current artifact identity. The accepted package and skill release is 0.35.3. Its instructions deliberately select the independently reviewed CLI 0.35.2; the release version and executable review pin are separate identities. Follow the host installation guide for exact tested versions and review steps.