/v1 is the contract. Everything else forwards to it and adds only what its audience needs: types for the SDKs, a terminal for the CLI, tools for an agent. An operation has the same name and the same meaning wherever it exists. A test checks this table against the OpenAPI document, both SDKs, the CLI's commands and the MCP tools, so a layer cannot gain, lose or rename one on its own.| operation | REST | TypeScript | Python | CLI | MCP tool |
| find | POST /v1/find | find | find | find | orsel_find |
| read | POST /v1/read | read | read | read | orsel_read |
| explore | POST /v1/explore | explore | explore | explore | orsel_explore |
| actions | GET /v1/actions | actions.list | actions.list | actions | orsel_actions |
| act | POST /v1/act | act | act | act | orsel_act |
| run | GET /v1/runs/{id} | runs.get | runs.get | runs | orsel_run |
| runs | GET /v1/runs | runs.page | runs.page | runs | orsel_runs |
| explain | GET /v1/runs/{id}/explain | explain | explain | explain | orsel_explain |
| approvals | GET /v1/approvals | approvals.page | approvals.page | approvals | orsel_approvals |
| approve | POST /v1/approvals/{id}/approve | approve | approve | approve | orsel_approve |
| deny | POST /v1/approvals/{id}/deny | deny | deny | deny | orsel_approve |
| revoke | POST /v1/revoke | revoke | revoke | revoke | orsel_revoke |
| sites | GET /v1/sites | sites.page | sites.page | sites | orsel_sites |
| recipes | GET /v1/recipes | recipes.page | recipes.page | recipes | orsel_recipes |
| policy | GET /v1/policy | policy.get | policy.get | policy | orsel_policy |
| policyCheck | POST /v1/policy/dry-run | policy.dryRun | policy.dry_run | policy | orsel_policy |
| ledgerVerify | GET /v1/ledger/verify | ledger.verify | ledger.verify | verify | orsel_ledger_verify |
| benchmark | GET /v1/benchmark | benchmark.get | benchmark.get | bench | orsel_benchmark |
| usage | GET /v1/usage | usage | usage | usage | orsel_usage |
| whoami | GET /v1/whoami | whoami | whoami | whoami | orsel_whoami |
| watch | POST /v1/watches | watches.create | watches.create | watch | orsel_watch |
| watches | GET /v1/watches | watches.list | watches.list | watch | orsel_watches |
| unwatch | DELETE /v1/watches/{id} | watches.delete | watches.delete | watch | orsel_unwatch |
site(domain) in both SDKs is actions plus act, one method per action (use a website as an API). typegen in the CLI reads actions and writes types.Idempotency-Key on writes, one error envelope, an OpenAPI document at /v1/openapi.json.site().--json, exit codes). act takes --set key=value.orsel is enough to start and says which to load next: orsel-explore (turn a website into an API), orsel-approvals (writes, forms, secrets), orsel-sdk (TypeScript and Python, site(), typegen, errors), orsel-mcp (the tool catalog and what a model may do) and orsel-operate (runs, ledger, usage, reliability, watches, policy).123npx skills add egeuysall/orsel-skills # the public repository that holds only the skills npx skills add ./ --list # from a checkout: see what it finds orsel init # writes them into a project, with the MCP config and rules
/.well-known/agent-skills/index.json for agents that discover skills over HTTP. The files live in skills/<name>/SKILL.md and are generated from one source (bun run skills:gen); bun run skills:check fails when they are out of date, and bun run skills:sync copies them to the public repository.