How long does the integration take?

Change one base URL in an OpenAI-compatible client, or install the Python or TypeScript SDK; the quickstart routes a first request in a few minutes.

Will this lock us into another vendor?

No. The router is Apache-2.0 open source, your provider keys stay yours, and the API is OpenAI-compatible, so leaving is a change of base URL.

What happens when a provider fails?

The router falls back to the next model your policy allows and records the fallback in the trace, so an outage costs a retry, not an incident.

data - Skill - OpenSmartRoute

Imported from docxology/bond (data/AGENTS.md). Install upstream with npx skills add docxology/bond --skill data. Copyright stays with the author.

Data Directory — Agent Guide

Versioned project inputs only. Pipeline outputs must not be committed here.

`claim_ledger.yaml`

Evidence-registry for manuscript claims that are intentionally sourced from code, examples, or generated reports rather than {{VARIABLE}} injection.

The compendium declares no rows: every number it prints is either imported from a source package's manuscript or injected as a {{TOKEN}} from live discovery (src/bond_manuscript/manuscript_variables.py). The file is kept for the schema, so a future claim that genuinely needs a ledger has a place to go.

Schema (preserve when adding rows)

Field	Purpose
`claim_id`	Stable identifier
`kind`	Claim category
`value`	Declared numeric or textual value
`source`	Provenance (module, manuscript section, artifact)
`source_tier`	Trust tier for validation
`freshness`	Staleness policy
`artifact_path`	Optional path to backing file

Edit protocol

Edit only when manuscript claims, figure defaults, or source-backed numeric facts change.
Re-run evidence validation / pipeline stages that consume the ledger.
Do not store generated CSV/JSON/PNG under data/.

Quick orientation: README.md.

Install data 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.

data

About

Data Directory — Agent Guide

`claim_ledger.yaml`

Schema (preserve when adding rows)

Edit protocol

Use it

Install data and it becomes one more candidate the router can pick - when it fits.

What it costs to adopt - in engineering time.

One base URL

No vendor lock-in

Failover built in

Manifest

Reviews