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.

structural-audit - Prompt - OpenSmartRoute

Name: structural-audit
Author: JackSmack1971

Prompt file imported from JackSmack1971/ai-trading-arena (.claude/commands/structural-audit.md). Copyright stays with the author.

Run the deterministic Claude-framework audit for this repository.

Execute node .claude/workflows/arena-audit.js from the repository root.
Treat every reported finding as a framework-surface defect until verified otherwise.
Confirm the audit verified hook registration and hook script presence rather than only file counts.
Confirm the audit verified the .claude/worktrees/ runtime parent when the framework references isolated worktrees.
Confirm the audit verified destructive deny coverage in .claude/settings.json, including the PowerShell-native variants used in this workspace.
Confirm the audit verified the project-level security guidance surfaces if the framework relies on simulator-boundary and secret-handling review criteria.
Confirm the audit verified output-style frontmatter so .claude/output-styles/*.md files remain real Claude output-style artifacts rather than loose markdown notes.
Confirm the audit verified frontmatter shape for .claude/commands/*.md, .claude/output-styles/*.md, .claude/agents/*.md, .claude/skills/*/SKILL.md, and .claude/rules/*.md.
Confirm the audit verified artifact identity consistency so rule, agent, output-style, and skill name metadata matches the owning filename or directory.
Confirm the audit verified that backticked framework evidence paths in docs/FRAMEWORK_TRACEABILITY.md still resolve.
If the workflow is clean, spot-check CLAUDE.md, AGENTS.md, .claude/settings.json, .claude/claude-security-guidance.md, .claude/security-patterns.json, .claude/hooks/, .claude/worktrees/, and the most relevant rule files for stale or conflicting guidance.
Interpret artifact counts as reusable framework surfaces only; support docs such as README.md do not count toward command, agent, rule, or output-style totals.
Report the result with counts, evidence paths, and any remaining content-level cleanup opportunities.

Install structural-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.

{ "prompt": "Run the deterministic Claude-framework audit for this repository.\n\n1. Execute `node .claude/workflows/arena-audit.js` from the repository root.\n2. Treat every reported finding as a framework-surface defect until verified otherwise.\n3. Confirm the audit verified hook registration and hook script presence rather than only file counts.\n4. Confirm the audit verified the `.claude/worktrees/` runtime parent when the framework references isolated worktrees.\n5. Confirm the audit verified destructive deny coverage in `.claude/settings.json`, including the PowerShell-native variants used in this workspace.\n6. Confirm the audit verified the project-level security guidance surfaces if the framework relies on simulator-boundary and secret-handling review criteria.\n7. Confirm the audit verified output-style frontmatter so `.claude/output-styles/*.md` files remain real Claude output-style artifacts rather than loose markdown notes.\n8. Confirm the audit verified frontmatter shape for `.claude/commands/*.md`, `.claude/output-styles/*.md`, `.claude/agents/*.md`, `.claude/skills/*/SKILL.md`, and `.claude/rules/*.md`.\n9. Confirm the audit verified artifact identity consistency so rule, agent, output-style, and skill `name` metadata matches the owning filename or directory.\n10. Confirm the audit verified that backticked framework evidence paths in `docs/FRAMEWORK_TRACEABILITY.md` still resolve.\n11. If the workflow is clean, spot-check `CLAUDE.md`, `AGENTS.md`, `.claude/settings.json`, `.claude/claude-security-guidance.md`, `.claude/security-patterns.json`, `.claude/hooks/`, `.claude/worktrees/`, and the most relevant rule files for stale or conflicting guidance.\n12. Interpret artifact counts as reusable framework surfaces only; support docs such as `README.md` do not count toward command, agent, rule, or output-style totals.\n13. Report the result with counts, evidence paths, and any remaining content-level cleanup opportunities.", "variables": [], "notes": "Run the deterministic structural audit for Claude-native framework surfaces. Tools: Read, Bash.", "metadata": { "source": { "provider": "github-claude-commands", "repository": "https://github.com/JackSmack1971/ai-trading-arena", "path": ".claude/commands/structural-audit.md", "ref": "4856bd9b8a83284ea757db8a1e37babf8786175f", "url": "https://github.com/JackSmack1971/ai-trading-arena/blob/4856bd9b8a83284ea757db8a1e37babf8786175f/.claude/commands/structural-audit.md", "key": "JackSmack1971/ai-trading-arena/.claude/commands/structural-audit.md" } } }

structural-audit

About

Install structural-audit 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

Use it

Manifest

Reviews