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.

development - Skill - OpenSmartRoute

Name: development
Author: ActiveInferenceInstitute

Imported from ActiveInferenceInstitute/cognitive (docs/development/AGENTS.md). Install upstream with npx skills add ActiveInferenceInstitute/cognitive --skill development. Copyright stays with the author.

Development documentation instructions

The development documentation records how to work on this repository: the development loop, quality gates, and conventions. It does not describe automated development agents.

Development loop

python -m pip install -e ".[dev]"
python -m pytest -q --cov
ruff check .
mypy code/tools/src code/Things code/scripts
python -m compileall -q code
python code/scripts/validate_docs.py --json
python code/scripts/verify_links.py . --json

The complete loop, including the manuscript build, benchmark, and mutation check, is in docs/development/README.md.

Conventions

Python: snake_case, type hints, docstrings on public methods, ruff for lint and format (project style, no black).
Documentation: every content folder has an AGENTS.md (instructions) and a README.md (overview/navigation); YAML frontmatter is required on Markdown files; internal cross-references use [[wiki links]].
Tests: real data and computation, no fixture-only assertions; tests write artifacts only to temporary directories.

Rules

Update docs/development/README.md when the tooling or gate commands change.
Do not document fictional development workflows or agent frameworks; the repository's actual tooling lives in code/scripts/ and the installed CLI entry points.

Install development 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.

{ "ocm": "1", "id": "activeinferenceinstitute-cognitive-development", "kind": "skill", "name": "development", "description": "The development documentation records how to work on this repository: the development loop, quality gates, and conventions. It does not describe automated development agents.", "publisher": "ActiveInferenceInstitute", "version": "1.0.0", "capabilities": { "domains": [ "coding" ], "tags": [ "skill-md", "development", "agents", "workflow", "collaboration", "github-agents" ], "languages": [ "en" ] }, "quality_prior": 0.6, "examples": [ "The development documentation records how to work on this repository: the development loop, quality gates, and conventions. It does not describe automated development agents." ], "primary": false, "metadata": { "source": { "provider": "github-agents", "repository": "https://github.com/ActiveInferenceInstitute/cognitive", "path": "docs/development/AGENTS.md", "ref": "109ff3928fe0b21c6a78a91d04db763d4f238fa1", "url": "https://github.com/ActiveInferenceInstitute/cognitive/blob/109ff3928fe0b21c6a78a91d04db763d4f238fa1/docs/development/AGENTS.md", "key": "ActiveInferenceInstitute/cognitive/docs/development/AGENTS.md" } }, "instructions": "# Development documentation instructions\n\nThe development documentation records how to work on this repository: the\ndevelopment loop, quality gates, and conventions. It does not describe\nautomated development agents.\n\n## Development loop\n\n```bash\npython -m pip install -e \".[dev]\"\npython -m pytest -q --cov\nruff check .\nmypy code/tools/src code/Things code/scripts\npython -m compileall -q code\npython code/scripts/validate_docs.py --json\npython code/scripts/verify_links.py . --json\n```\n\nThe complete loop, including the manuscript build, benchmark, and mutation\ncheck, is in `docs/development/README", "cost": { "context_tokens": 333 } }

development

About

Development documentation instructions

Development loop

Conventions

Rules

Install development 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