Show HN: Self-hostable dashboard to govern a team's AI coding spend

Description and use

**Brief Summary (2‑3 sentences)**
Show HN is a self‑hostable web dashboard that lets teams monitor and control the cost of AI‑powered coding tools (e.g., code‑completion, RAG, or agent workflows). By plugging into existing model APIs and usage logs, it provides real‑time spend visualisations, budgeting alerts, and per‑project breakdowns, helping teams add AI capabilities without building a cost‑tracking stack from scratch.

**Value**
The dashboard turns opaque API‑usage data into actionable insights, enabling engineering managers to enforce budgets, compare providers, and justify AI investments. It accelerates prototyping of AI features—such as code‑completion, retrieval‑augmented generation, or autonomous agents—by surfacing spend early, so teams can iterate confidently while avoiding surprise bills.

**Practical Adoption Path**

1. **Pilot Setup** – Clone the repo, run the Docker compose file, and point the collector to the API keys/log endpoints of the LLM services you already use.
2. **Data Ingestion** – Configure the minimal “integration signals” (e.g., request logs, token counts) required for the dashboard; the project currently expects manual mapping, so verify that your logging format matches the sample schema.
3. **Validation** – Compare the dashboard’s spend figures against your cloud provider’s billing console for a week to confirm accuracy.
4. **Policy Layer** – Add budget thresholds and alert hooks (Slack, email) using the built‑in webhook configuration.
5. **Roll‑out** – Deploy the dashboard in your internal Kubernetes or VM environment and grant read‑only access to developers and finance stakeholders.

**Production Readiness**
The project is at a *medium* maturity level: it is recent (last updated 2026‑07‑02) and functional for internal prototypes, but the integration signals are sparse and the documentation is thin. Before production use, you should:

• Verify the open‑source license and any third‑party dependencies.
• Review the issue tracker for unresolved bugs and the release cadence to gauge long‑term maintenance.
• Harden the deployment (TLS, auth, network isolation) and set up monitoring for the dashboard itself.

If those checks pass, the dashboard can be safely used for internal cost governance; for mission‑critical, high‑scale environments you may want to supplement it with additional observability tooling or contribute enhancements back to the project.

Use in a workflow

Show HN: Self-hostable dashboard to govern a team's AI coding spend helps add AI capability without starting from a blank model stack.

• prototype AI features
• build RAG or agent workflows
• evaluate model tooling

Limitations and review notes

Medium: useful for prototypes or internal workflows, with dependency and maintenance checks before production.