@elevasis/sdk 0.5.12 → 0.5.13
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/dist/cli.cjs +92 -81
- package/dist/index.d.ts +11 -253
- package/dist/index.js +5 -9
- package/dist/templates.js +26 -23
- package/dist/types/worker/adapters/index.d.ts +0 -1
- package/dist/worker/index.js +47 -53
- package/package.json +1 -1
- package/reference/_navigation.md +13 -57
- package/reference/concepts.mdx +203 -0
- package/reference/deployment/{command-center-ui.mdx → command-center.mdx} +229 -151
- package/reference/deployment/index.mdx +158 -153
- package/reference/framework/agent.mdx +168 -151
- package/reference/framework/index.mdx +182 -103
- package/reference/framework/memory.mdx +347 -347
- package/reference/framework/tutorial-system.mdx +253 -0
- package/reference/{getting-started/index.mdx → getting-started.mdx} +4 -4
- package/reference/index.mdx +117 -114
- package/reference/platform-tools/adapters.mdx +175 -32
- package/reference/platform-tools/index.mdx +354 -195
- package/reference/resources/index.mdx +5 -0
- package/reference/{roadmap/index.mdx → roadmap.mdx} +1 -1
- package/reference/{runtime/index.mdx → runtime.mdx} +196 -141
- package/dist/types/worker/adapters/trello.d.ts +0 -14
- package/reference/concepts/index.mdx +0 -203
- package/reference/deployment/command-view.mdx +0 -154
- package/reference/framework/documentation.mdx +0 -92
- package/reference/platform-tools/examples.mdx +0 -170
- package/reference/runtime/limits.mdx +0 -75
- package/reference/security/credentials.mdx +0 -141
- /package/reference/{cli/index.mdx → cli.mdx} +0 -0
- /package/reference/{developer → framework}/interaction-guidance.mdx +0 -0
- /package/reference/{troubleshooting/common-errors.mdx → troubleshooting.mdx} +0 -0
|
@@ -0,0 +1,253 @@
|
|
|
1
|
+
---
|
|
2
|
+
title: Tutorial System
|
|
3
|
+
description: Reference for the SDK tutorial system -- 21-item menu across 4 sections, skill-level adaptation rules, progress tracking format, and per-lesson module contents
|
|
4
|
+
---
|
|
5
|
+
|
|
6
|
+
**Implementation:** `packages/sdk/src/cli/commands/templates/core/claude.ts` (`claudeTutorialCommandTemplate()`)
|
|
7
|
+
|
|
8
|
+
**Template Version:** 27
|
|
9
|
+
|
|
10
|
+
The tutorial system is emitted to `external/<project>/.claude/commands/tutorial.md` during `elevasis-sdk init` or `elevasis-sdk update`. Users invoke it with `/tutorial` from Claude Code.
|
|
11
|
+
|
|
12
|
+
---
|
|
13
|
+
|
|
14
|
+
## Menu System
|
|
15
|
+
|
|
16
|
+
`/tutorial` always shows the full 21-item menu table on invocation. All items are visible at once with progress indicators. No track gating -- any item is available at any time.
|
|
17
|
+
|
|
18
|
+
```
|
|
19
|
+
Elevasis Tutorial
|
|
20
|
+
==================
|
|
21
|
+
|
|
22
|
+
INTRODUCTION 0/4
|
|
23
|
+
1 Welcome & Orientation [ ]
|
|
24
|
+
2 How This Workspace Works [ ]
|
|
25
|
+
3 The /meta Command [ ]
|
|
26
|
+
4 /work and /docs [ ]
|
|
27
|
+
|
|
28
|
+
CORE CONCEPTS 0/6
|
|
29
|
+
5 Your First Custom Workflow [ ]
|
|
30
|
+
6 Understanding Data (Schemas) [ ]
|
|
31
|
+
7 Using Platform Tools [ ]
|
|
32
|
+
8 Multi-Step Workflows [ ]
|
|
33
|
+
9 Decision Points [ ]
|
|
34
|
+
10 Going to Production [ ]
|
|
35
|
+
|
|
36
|
+
ADVANCED MODULES 0/9
|
|
37
|
+
11 Human-in-the-Loop [ ]
|
|
38
|
+
12 Task Scheduling [ ]
|
|
39
|
+
13 Notification System [ ]
|
|
40
|
+
14 Real-World Integrations [ ]
|
|
41
|
+
15 Error Handling Mastery [ ]
|
|
42
|
+
16 Advanced Workflows [ ]
|
|
43
|
+
17 Resource Composition [ ]
|
|
44
|
+
18 LLM Integration [ ]
|
|
45
|
+
19 AI Agents [ ]
|
|
46
|
+
|
|
47
|
+
ADVANCED WORKSPACE 0/2
|
|
48
|
+
20 Rules, Memory, and Customization [ ]
|
|
49
|
+
21 Template Lifecycle [ ]
|
|
50
|
+
|
|
51
|
+
Pick a number to start.
|
|
52
|
+
```
|
|
53
|
+
|
|
54
|
+
**Progress indicators:** `[ ]` not started, `[>]` in progress, `[x] YYYY-MM-DD` complete.
|
|
55
|
+
|
|
56
|
+
User picks a number to start or resume any item. `"status"` shows the same table.
|
|
57
|
+
|
|
58
|
+
---
|
|
59
|
+
|
|
60
|
+
## 4 Sections
|
|
61
|
+
|
|
62
|
+
- **INTRODUCTION (1-4):** Welcome + workspace orientation (workspace framework, /meta, /work and /docs) -- taught before building
|
|
63
|
+
- **CORE CONCEPTS (5-10):** Orchestration lessons (workflows, schemas, tools, multi-step, conditions, production)
|
|
64
|
+
- **ADVANCED MODULES (11-19):** Hands-on feature modules (hitl, schedules, notifications, integrations, error-handling, workflows, composition, llm, agents)
|
|
65
|
+
- **ADVANCED WORKSPACE (20-21):** Power-user customization (rules/memory, template lifecycle)
|
|
66
|
+
|
|
67
|
+
---
|
|
68
|
+
|
|
69
|
+
## Skill-Level Adaptation
|
|
70
|
+
|
|
71
|
+
Assessed during `/meta init`. Two independent dimensions.
|
|
72
|
+
|
|
73
|
+
### automation dimension
|
|
74
|
+
|
|
75
|
+
| Level | Context Loading | Delivery | Verification |
|
|
76
|
+
| ---------- | ------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------- | ----------------------------------------------------------------- |
|
|
77
|
+
| `none` | Glossary, Workflow overview, Platform Tools Overview only. Skip Zod, Execution Model, Design Decisions. | Agent writes all code. No code shown to user. Analogies throughout. | CLI primary. Agent runs `elevasis-sdk exec` and narrates results. |
|
|
78
|
+
| `low-code` | All concept sections. Zapier/Make mapping. `adapters.mdx` on-demand. | Map to Zapier/Make equivalents. Show code with explanations. | CLI primary. |
|
|
79
|
+
| `custom` | All docs listed per-lesson and per-module. | Full technical content, code-first. | CLI primary. |
|
|
80
|
+
|
|
81
|
+
### platformNavigation dimension
|
|
82
|
+
|
|
83
|
+
| Level | Behavior |
|
|
84
|
+
| ------------- | ---------------------------------------------------------------------------------------------------------------------- |
|
|
85
|
+
| `none` | Walk through pages step by step with exact navigation paths. Explain each page before asking user to interact with it. |
|
|
86
|
+
| `oriented` | Reference pages by name; briefly remind purpose on first mention. |
|
|
87
|
+
| `comfortable` | Reference by name only; focus on advanced filtering and schedule types. |
|
|
88
|
+
|
|
89
|
+
**General rules (all levels):** Fast users: acknowledge, offer to skip ahead. Level change observed: note in `skills.md` Growth Log. After lesson/module: update progress file.
|
|
90
|
+
|
|
91
|
+
---
|
|
92
|
+
|
|
93
|
+
## Progress Logic
|
|
94
|
+
|
|
95
|
+
```
|
|
96
|
+
1. Read .claude/memory/tutorial-progress.md
|
|
97
|
+
2. Show the full menu table with progress indicators filled in
|
|
98
|
+
3. User picks a number -> start or resume that item directly
|
|
99
|
+
4. After completion: mark done in progress file, show updated menu table
|
|
100
|
+
5. All 21 items complete -> congratulate, suggest docs/ or /work
|
|
101
|
+
```
|
|
102
|
+
|
|
103
|
+
---
|
|
104
|
+
|
|
105
|
+
## Introduction: Items 1-4
|
|
106
|
+
|
|
107
|
+
Items 1-4 orient the user to the platform and workspace before any building.
|
|
108
|
+
|
|
109
|
+
**Item 1 -- Welcome & Orientation:** Platform tour, deploy echo workflow, Command Center orientation.
|
|
110
|
+
|
|
111
|
+
**Item 2 -- How This Workspace Works:** CLAUDE.md as instruction sheet, four commands, creds skill, memory system overview.
|
|
112
|
+
|
|
113
|
+
**Item 3 -- The /meta Command:** Project dashboard, /meta fix, /meta deploy, /meta health.
|
|
114
|
+
|
|
115
|
+
**Item 4 -- /work and /docs:** Cross-session task tracking, documentation lifecycle, /work vs /docs boundary.
|
|
116
|
+
|
|
117
|
+
---
|
|
118
|
+
|
|
119
|
+
## Core Path: Lessons 5-10
|
|
120
|
+
|
|
121
|
+
Each lesson: (1) announce title and goal, (2) explain concept per skill level, (3) build or modify something, (4) verify it works, (5) record observations, ask "ready for next?"
|
|
122
|
+
|
|
123
|
+
**Lesson 5 -- Your First Custom Workflow**
|
|
124
|
+
|
|
125
|
+
- `automation: none` -- "A recipe." Agent writes all code. Agent runs `elevasis-sdk exec` and narrates result.
|
|
126
|
+
- `automation: low-code or custom` -- Modify echo workflow. Walk through `config`, `contract`, `steps`, `entryPoint`. Deploy with `elevasis-sdk check` then `elevasis-sdk deploy`.
|
|
127
|
+
|
|
128
|
+
**Lesson 6 -- Understanding Data (Schemas)**
|
|
129
|
+
|
|
130
|
+
- `automation: none` -- "What information does your automation need?" No Zod, no code shown. Agent reads identity.md goals and writes the schema.
|
|
131
|
+
- `automation: low-code` -- "Field mapping like Zapier, but with validation." Show Zod types briefly.
|
|
132
|
+
- `automation: custom` -- Full Zod explanation. `z.infer`. Show how schema fields define CLI input structure.
|
|
133
|
+
|
|
134
|
+
**Lesson 7 -- Using Platform Tools**
|
|
135
|
+
|
|
136
|
+
- `automation: none` -- Agent makes all code edits. Credential creation guided via Command Center UI.
|
|
137
|
+
- `automation: low-code or custom` -- Explain adapter pattern: `createAttioAdapter('cred')`. Show singleton pattern: `import { scheduler, llm } from '@elevasis/sdk/worker'`.
|
|
138
|
+
|
|
139
|
+
**Lesson 8 -- Multi-Step Workflows**
|
|
140
|
+
|
|
141
|
+
- `automation: none` -- "Step 1 passes its result to Step 2, like a relay race." Command View is PRIMARY teaching tool.
|
|
142
|
+
- `automation: low-code or custom` -- Chain steps with `StepType.LINEAR`. Show relationship edges in Command View.
|
|
143
|
+
|
|
144
|
+
**Lesson 9 -- Decision Points**
|
|
145
|
+
|
|
146
|
+
- `automation: none` -- "If the customer is VIP, do this -- otherwise, do that." Agent runs both paths. Open Execution Logs.
|
|
147
|
+
- `automation: low-code or custom` -- Conditional routing with `StepType.CONDITIONAL`. Show step trace in Execution Logs.
|
|
148
|
+
|
|
149
|
+
**Lesson 10 -- Going to Production**
|
|
150
|
+
|
|
151
|
+
- `automation: none` -- "Draft vs live." Create Recurring schedule in Task Scheduler.
|
|
152
|
+
- `automation: low-code or custom` -- Change `status: dev` -> `production`. Cover `try/catch`, `ExecutionError`, `PlatformToolError`. CLI monitoring: `elevasis-sdk executions`.
|
|
153
|
+
|
|
154
|
+
---
|
|
155
|
+
|
|
156
|
+
## Module System: Items 11-19
|
|
157
|
+
|
|
158
|
+
Modules are available any time -- no core path prerequisite. After completing a module the full menu table is shown again.
|
|
159
|
+
|
|
160
|
+
| # | Module | Reference Docs | Build | Key Concepts |
|
|
161
|
+
| --- | ---------------- | ---------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------ |
|
|
162
|
+
| 11 | `hitl` | `command-center.mdx` (Command Queue) | Approval gate with `approval.requestApproval()`. Full lifecycle: trigger -> Command Queue -> approve/reject -> resume. | `approval` adapter, pending state, Command Queue UI |
|
|
163
|
+
| 12 | `schedules` | `command-center.mdx` (Task Scheduler) | All three schedule types: Recurring, Relative, Absolute. `scheduler` adapter for in-workflow. | Cron syntax, schedule types, Task Scheduler UI |
|
|
164
|
+
| 13 | `notifications` | `adapters.mdx` (notifications, email) | Notification + email steps. Alerts on completion. | `notifications` singleton, `email` singleton |
|
|
165
|
+
| 14 | `integrations` | `platform-tools/index.mdx` (Credential Security), `adapters.mdx` | Real adapter from `identity.md`. Full end-to-end with OAuth or API key credential. | Adapter pattern, credential scoping, external call error handling |
|
|
166
|
+
| 15 | `error-handling` | `patterns.mdx` (error handling), `troubleshooting.mdx` | Workflow demonstrating all three error types. `try/catch`, `context.logger`, recovery. | `ExecutionError`, `PlatformToolError`, `ToolingError` |
|
|
167
|
+
| 16 | `workflows` | `patterns.mdx` (store, logging, organization) | Refactor with `context.store`, `context.logger`, domain directories. Advanced schema patterns. | `context.store`, `context.logger`, schema depth |
|
|
168
|
+
| 17 | `composition` | `command-center.mdx` (Command View), `patterns.mdx` | Two workflows: first triggers second with `execution.trigger()`. Declare relationship. | `execution.trigger`, relationship declarations, Command View edges |
|
|
169
|
+
| 18 | `llm` | `adapters.mdx` (llm singleton) | `llm.generate()` with structured output. Model selection and temperature. | `llm` singleton, structured output, temperature |
|
|
170
|
+
| 19 | `agents` | `reference/framework/agent.mdx` | Agent definition with tools. LLM tool calling. Agent vs workflow comparison. | Agent definition, tool registration, execution trace |
|
|
171
|
+
|
|
172
|
+
---
|
|
173
|
+
|
|
174
|
+
## Advanced Workspace: Items 20-21
|
|
175
|
+
|
|
176
|
+
**Item 20 -- Rules, Memory, and Customization**
|
|
177
|
+
|
|
178
|
+
- `none` -- "Sticky notes the agent reads automatically." Show `.claude/rules/workspace-patterns.md` without technical detail.
|
|
179
|
+
- `low-code` -- Show `.claude/rules/` directory. Explain MANAGED (`sdk-patterns.md`) vs INIT_ONLY (`workspace-patterns.md`). Explain error promotion: 3+ recurrences -> add a rule.
|
|
180
|
+
- `custom` -- Read both rules files. Walk through memory layout: `index.md`, `profile/`, `errors/`, `tutorial-progress.md`. Walk through adding a new rule.
|
|
181
|
+
|
|
182
|
+
**Item 21 -- Template Lifecycle**
|
|
183
|
+
|
|
184
|
+
- `none` -- "SDK releases automatically update your workspace." Explain `/meta fix` as the update command.
|
|
185
|
+
- `low-code` -- Explain MANAGED vs INIT_ONLY file types. Show how `/meta fix` handles conflicts.
|
|
186
|
+
- `custom` -- Full template lifecycle: init (writes all files), update (applies MANAGED changes, flags conflicts, preserves INIT_ONLY), fix (drift repair + SDK upgrade offer). Conflict resolution reads current + template from `@elevasis/sdk/templates`, merges preserving customizations.
|
|
187
|
+
|
|
188
|
+
---
|
|
189
|
+
|
|
190
|
+
## Progress File Format
|
|
191
|
+
|
|
192
|
+
Stored at `.claude/memory/tutorial-progress.md`.
|
|
193
|
+
|
|
194
|
+
```markdown
|
|
195
|
+
# Tutorial Progress
|
|
196
|
+
|
|
197
|
+
Current: <free-form, e.g. "4: /work and /docs" or "M:integrations">
|
|
198
|
+
Started: YYYY-MM-DD
|
|
199
|
+
Last Session: YYYY-MM-DD
|
|
200
|
+
|
|
201
|
+
## Completed Lessons
|
|
202
|
+
|
|
203
|
+
| Item | Title | Completed | Duration |
|
|
204
|
+
| --- | --- | --- | --- |
|
|
205
|
+
| 1 | Welcome & Orientation | 2026-03-01 | 25 min |
|
|
206
|
+
|
|
207
|
+
## Completed Modules
|
|
208
|
+
|
|
209
|
+
| Module | Title | Completed | Duration |
|
|
210
|
+
| --- | --- | --- | --- |
|
|
211
|
+
| hitl | Human-in-the-Loop | 2026-03-03 | 40 min |
|
|
212
|
+
|
|
213
|
+
## Capability Observations
|
|
214
|
+
|
|
215
|
+
| Source | Observation |
|
|
216
|
+
| --- | --- |
|
|
217
|
+
| 1 | Navigated Command View without guidance |
|
|
218
|
+
|
|
219
|
+
## Assessment Notes
|
|
220
|
+
|
|
221
|
+
- Strong automation background, slow on TypeScript syntax
|
|
222
|
+
```
|
|
223
|
+
|
|
224
|
+
**Backward compatibility:** Missing `Completed Modules` section treated as empty. Old files with `Completed MF Lessons` map entries to items 2-4 and 20-21 by title match.
|
|
225
|
+
|
|
226
|
+
---
|
|
227
|
+
|
|
228
|
+
## Implementation Files
|
|
229
|
+
|
|
230
|
+
| File | Role |
|
|
231
|
+
| --------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
|
232
|
+
| `packages/sdk/src/cli/commands/templates/core/claude.ts` | `claudeTutorialCommandTemplate()` -- full tutorial prompt; `claudeMdTemplate()` -- Platform Navigation guidance; `claudeMetaCommandTemplate()` -- onboarding assessment |
|
|
233
|
+
| `packages/sdk/src/cli/commands/templates/core/workspace.ts` | `TEMPLATE_VERSION = 27` |
|
|
234
|
+
| `packages/sdk/src/cli/commands/init.ts` | `TEMPLATE_CHANGELOG` -- version history (v1-v27) |
|
|
235
|
+
| `apps/docs/content/docs/sdk/framework/interaction-guidance.mdx` | Skill dimensions: automation, platformNavigation |
|
|
236
|
+
|
|
237
|
+
---
|
|
238
|
+
|
|
239
|
+
## Template Version History (Recent)
|
|
240
|
+
|
|
241
|
+
| Version | Change |
|
|
242
|
+
| ------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------ |
|
|
243
|
+
| 27 | 4-section menu redesign: INTRODUCTION (1-4), CORE CONCEPTS (5-10), ADVANCED MODULES (11-19), ADVANCED WORKSPACE (20-21); MF1-MF3 moved to front as items 2-4 |
|
|
244
|
+
| 23 | Flat 21-item menu table; module gating removed; error-handling moved to position 12 |
|
|
245
|
+
| 22 | `/docs` command re-introduced; MF3 covers `/work` and `/docs` together |
|
|
246
|
+
| 20 | `/work` comprehensive rewrite -- list-then-pick, interview-based create, intelligent placement |
|
|
247
|
+
| 19 | Menu-driven entry; Meta-Framework track (MF1-MF5) with skill-adaptive variants |
|
|
248
|
+
| 18 | Skill-adaptive lesson variants; Command Center UI integrated; Platform Navigation dimension |
|
|
249
|
+
| 16 | Module system; single `/tutorial` invocation; phase tracking |
|
|
250
|
+
|
|
251
|
+
---
|
|
252
|
+
|
|
253
|
+
**Last Updated:** 2026-03-05
|
|
@@ -136,10 +136,10 @@ Replace `<execution-id>` with the ID returned from the executions list.
|
|
|
136
136
|
## Next Steps
|
|
137
137
|
|
|
138
138
|
- **Open in Claude Code** -- Setup runs automatically on first session (installs deps, configures `.env`, creates your developer profile)
|
|
139
|
-
- [Development Framework](
|
|
140
|
-
- [Resources](
|
|
141
|
-
- [CLI Reference](
|
|
142
|
-
- [Deployment](
|
|
139
|
+
- [Development Framework](framework/index.mdx) -- How Claude Code helps you build
|
|
140
|
+
- [Resources](resources/index.mdx) -- Define workflows and agents
|
|
141
|
+
- [CLI Reference](cli.mdx) -- Full command reference with flags
|
|
142
|
+
- [Deployment](deployment/index.mdx) -- Deployment lifecycle and environment variables
|
|
143
143
|
|
|
144
144
|
When a new SDK version is released, run `/meta fix` in Claude Code for an interactive upgrade that includes SDK update, drift repair, and documentation verification. Alternatively, run `elevasis-sdk update` from the terminal to update managed files (CLAUDE.md, slash commands, `.gitignore`) without agent interaction. To update the SDK package itself, run `pnpm update @elevasis/sdk`.
|
|
145
145
|
|
package/reference/index.mdx
CHANGED
|
@@ -1,114 +1,117 @@
|
|
|
1
|
-
---
|
|
2
|
-
title: Elevasis SDK
|
|
3
|
-
description: Build and deploy workflows, agents, and resources with the Elevasis SDK
|
|
4
|
-
loadWhen: "First session or new to the SDK"
|
|
5
|
-
---
|
|
6
|
-
|
|
7
|
-
`@elevasis/sdk` lets you build workflows, agents, and resources in TypeScript and deploy them to the Elevasis platform with a single command. The developer experience is Vercel-style: write TypeScript, validate locally, deploy -- the platform handles execution, tool access, and observability. You never manage infrastructure. Zod 4.1 is the only peer dependency.
|
|
8
|
-
|
|
9
|
-
## Quick Start
|
|
10
|
-
|
|
11
|
-
```bash
|
|
12
|
-
pnpm dlx @elevasis/sdk init my-project
|
|
13
|
-
cd my-project
|
|
14
|
-
pnpm install
|
|
15
|
-
elevasis-sdk deploy
|
|
16
|
-
```
|
|
17
|
-
|
|
18
|
-
After `pnpm dlx @elevasis/sdk init`, your project is scaffolded with a working echo workflow, config file, TypeScript setup, and a `CLAUDE.md` that gives Claude Code full awareness of the SDK. After `elevasis-sdk deploy`, your resources appear in AI Studio and are available to run immediately.
|
|
19
|
-
|
|
20
|
-
Here is a minimal workflow definition:
|
|
21
|
-
|
|
22
|
-
```ts
|
|
23
|
-
import { WorkflowDefinition, OrganizationResources } from '@elevasis/sdk';
|
|
24
|
-
import { z } from 'zod';
|
|
25
|
-
|
|
26
|
-
const greetInput = z.object({ name: z.string() });
|
|
27
|
-
const greetOutput = z.object({ message: z.string() });
|
|
28
|
-
|
|
29
|
-
type GreetInput = z.infer<typeof greetInput>;
|
|
30
|
-
type GreetOutput = z.infer<typeof greetOutput>;
|
|
31
|
-
|
|
32
|
-
const greetWorkflow: WorkflowDefinition = {
|
|
33
|
-
id: 'greet',
|
|
34
|
-
name: 'Greet',
|
|
35
|
-
inputSchema: greetInput,
|
|
36
|
-
outputSchema: greetOutput,
|
|
37
|
-
steps: [
|
|
38
|
-
{
|
|
39
|
-
id: 'greet-step',
|
|
40
|
-
name: 'Build greeting',
|
|
41
|
-
handler: async (input: GreetInput): Promise<GreetOutput> => {
|
|
42
|
-
return { message: `Hello, ${input.name}!` };
|
|
43
|
-
},
|
|
44
|
-
next: null,
|
|
45
|
-
},
|
|
46
|
-
],
|
|
47
|
-
};
|
|
48
|
-
|
|
49
|
-
export const resources: OrganizationResources = {
|
|
50
|
-
workflows: [greetWorkflow],
|
|
51
|
-
};
|
|
52
|
-
```
|
|
53
|
-
|
|
54
|
-
## What You Can Build
|
|
55
|
-
|
|
56
|
-
- **Workflows** -- Step-based automation with typed inputs and outputs. Steps can be linear, conditional, or branching. Each step is a plain async function. Workflows are the primary resource type and are fully supported.
|
|
57
|
-
- **Agents** -- Autonomous AI resources with access to platform tools. Agents run in the worker runtime with full LLM access and platform tool support. Use `--async` when executing agents to avoid HTTP timeout limits on long-running runs.
|
|
58
|
-
|
|
59
|
-
## Platform Tools
|
|
60
|
-
|
|
61
|
-
Inside any workflow step, import `platform` from `@elevasis/sdk/worker` to call platform tools:
|
|
62
|
-
|
|
63
|
-
```ts
|
|
64
|
-
import { platform } from '@elevasis/sdk/worker';
|
|
65
|
-
|
|
66
|
-
const result = await platform.call({
|
|
67
|
-
tool: 'gmail',
|
|
68
|
-
method: 'send',
|
|
69
|
-
params: { to: 'user@example.com', subject: 'Hello', body: 'World' },
|
|
70
|
-
credential: 'my-gmail-credential',
|
|
71
|
-
});
|
|
72
|
-
```
|
|
73
|
-
|
|
74
|
-
The platform exposes 70+ tools across 13 integration adapters -- Gmail, Stripe, Google Sheets, Notion, and more. Credentials are managed server-side; API keys never cross the execution boundary. LLM calls are also available via `platform.call({ tool: 'llm', ... })` with API keys resolved from platform environment variables.
|
|
75
|
-
|
|
76
|
-
See [Platform Tools](platform-tools/index.mdx) for the full catalog.
|
|
77
|
-
|
|
78
|
-
## Known Limitations
|
|
79
|
-
|
|
80
|
-
- **No streaming logs** -- Execution logs are returned in the response body after completion. Real-time log streaming is not available.
|
|
81
|
-
- **Agent HTTP timeouts** -- Use `elevasis-sdk exec --async` for agent executions. Agents can run for minutes; the synchronous endpoint will time out for long-running runs. The `--async` flag returns an execution ID immediately and polls for the result.
|
|
82
|
-
|
|
83
|
-
## Documentation
|
|
84
|
-
|
|
85
|
-
### Getting Started
|
|
86
|
-
|
|
87
|
-
- [Getting Started](getting-started
|
|
88
|
-
|
|
89
|
-
### Core Concepts
|
|
90
|
-
|
|
91
|
-
- [Resources](resources/index.mdx) - Workflow and agent definition patterns, Zod schemas, step types, and routing
|
|
92
|
-
- [Platform Tools](platform-tools/index.mdx) - Full catalog of 70+ tools, integration adapters, and credential management
|
|
93
|
-
- [
|
|
94
|
-
|
|
95
|
-
### Reference
|
|
96
|
-
|
|
97
|
-
- [Concepts](concepts
|
|
98
|
-
- [Templates](templates/index.mdx) - 7 workflow templates: web-scraper, data-enrichment, email-sender, lead-scorer, and more
|
|
99
|
-
- [CLI Reference](cli
|
|
100
|
-
- [Deployment](deployment/index.mdx) - Deploy pipeline, versioning, bundle upload, and registry registration
|
|
101
|
-
- [Runtime](runtime
|
|
102
|
-
|
|
103
|
-
### Framework
|
|
104
|
-
|
|
105
|
-
- [Development Framework](framework/index.mdx) - How Claude Code helps you build: project structure, agent integration, memory, and documentation
|
|
106
|
-
|
|
107
|
-
|
|
108
|
-
|
|
109
|
-
|
|
110
|
-
|
|
111
|
-
|
|
112
|
-
|
|
113
|
-
|
|
114
|
-
|
|
1
|
+
---
|
|
2
|
+
title: Elevasis SDK
|
|
3
|
+
description: Build and deploy workflows, agents, and resources with the Elevasis SDK
|
|
4
|
+
loadWhen: "First session or new to the SDK"
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
`@elevasis/sdk` lets you build workflows, agents, and resources in TypeScript and deploy them to the Elevasis platform with a single command. The developer experience is Vercel-style: write TypeScript, validate locally, deploy -- the platform handles execution, tool access, and observability. You never manage infrastructure. Zod 4.1 is the only peer dependency.
|
|
8
|
+
|
|
9
|
+
## Quick Start
|
|
10
|
+
|
|
11
|
+
```bash
|
|
12
|
+
pnpm dlx @elevasis/sdk init my-project
|
|
13
|
+
cd my-project
|
|
14
|
+
pnpm install
|
|
15
|
+
elevasis-sdk deploy
|
|
16
|
+
```
|
|
17
|
+
|
|
18
|
+
After `pnpm dlx @elevasis/sdk init`, your project is scaffolded with a working echo workflow, config file, TypeScript setup, and a `CLAUDE.md` that gives Claude Code full awareness of the SDK. After `elevasis-sdk deploy`, your resources appear in AI Studio and are available to run immediately.
|
|
19
|
+
|
|
20
|
+
Here is a minimal workflow definition:
|
|
21
|
+
|
|
22
|
+
```ts
|
|
23
|
+
import { WorkflowDefinition, OrganizationResources } from '@elevasis/sdk';
|
|
24
|
+
import { z } from 'zod';
|
|
25
|
+
|
|
26
|
+
const greetInput = z.object({ name: z.string() });
|
|
27
|
+
const greetOutput = z.object({ message: z.string() });
|
|
28
|
+
|
|
29
|
+
type GreetInput = z.infer<typeof greetInput>;
|
|
30
|
+
type GreetOutput = z.infer<typeof greetOutput>;
|
|
31
|
+
|
|
32
|
+
const greetWorkflow: WorkflowDefinition = {
|
|
33
|
+
id: 'greet',
|
|
34
|
+
name: 'Greet',
|
|
35
|
+
inputSchema: greetInput,
|
|
36
|
+
outputSchema: greetOutput,
|
|
37
|
+
steps: [
|
|
38
|
+
{
|
|
39
|
+
id: 'greet-step',
|
|
40
|
+
name: 'Build greeting',
|
|
41
|
+
handler: async (input: GreetInput): Promise<GreetOutput> => {
|
|
42
|
+
return { message: `Hello, ${input.name}!` };
|
|
43
|
+
},
|
|
44
|
+
next: null,
|
|
45
|
+
},
|
|
46
|
+
],
|
|
47
|
+
};
|
|
48
|
+
|
|
49
|
+
export const resources: OrganizationResources = {
|
|
50
|
+
workflows: [greetWorkflow],
|
|
51
|
+
};
|
|
52
|
+
```
|
|
53
|
+
|
|
54
|
+
## What You Can Build
|
|
55
|
+
|
|
56
|
+
- **Workflows** -- Step-based automation with typed inputs and outputs. Steps can be linear, conditional, or branching. Each step is a plain async function. Workflows are the primary resource type and are fully supported.
|
|
57
|
+
- **Agents** -- Autonomous AI resources with access to platform tools. Agents run in the worker runtime with full LLM access and platform tool support. Use `--async` when executing agents to avoid HTTP timeout limits on long-running runs.
|
|
58
|
+
|
|
59
|
+
## Platform Tools
|
|
60
|
+
|
|
61
|
+
Inside any workflow step, import `platform` from `@elevasis/sdk/worker` to call platform tools:
|
|
62
|
+
|
|
63
|
+
```ts
|
|
64
|
+
import { platform } from '@elevasis/sdk/worker';
|
|
65
|
+
|
|
66
|
+
const result = await platform.call({
|
|
67
|
+
tool: 'gmail',
|
|
68
|
+
method: 'send',
|
|
69
|
+
params: { to: 'user@example.com', subject: 'Hello', body: 'World' },
|
|
70
|
+
credential: 'my-gmail-credential',
|
|
71
|
+
});
|
|
72
|
+
```
|
|
73
|
+
|
|
74
|
+
The platform exposes 70+ tools across 13 integration adapters -- Gmail, Stripe, Google Sheets, Notion, and more. Credentials are managed server-side; API keys never cross the execution boundary. LLM calls are also available via `platform.call({ tool: 'llm', ... })` with API keys resolved from platform environment variables.
|
|
75
|
+
|
|
76
|
+
See [Platform Tools](platform-tools/index.mdx) for the full catalog.
|
|
77
|
+
|
|
78
|
+
## Known Limitations
|
|
79
|
+
|
|
80
|
+
- **No streaming logs** -- Execution logs are returned in the response body after completion. Real-time log streaming is not available.
|
|
81
|
+
- **Agent HTTP timeouts** -- Use `elevasis-sdk exec --async` for agent executions. Agents can run for minutes; the synchronous endpoint will time out for long-running runs. The `--async` flag returns an execution ID immediately and polls for the result.
|
|
82
|
+
|
|
83
|
+
## Documentation
|
|
84
|
+
|
|
85
|
+
### Getting Started
|
|
86
|
+
|
|
87
|
+
- [Getting Started](getting-started.mdx) - Installation, authentication, first workflow, and project structure
|
|
88
|
+
|
|
89
|
+
### Core Concepts
|
|
90
|
+
|
|
91
|
+
- [Resources](resources/index.mdx) - Workflow and agent definition patterns, Zod schemas, step types, and routing
|
|
92
|
+
- [Platform Tools](platform-tools/index.mdx) - Full catalog of 70+ tools, integration adapters, and credential management
|
|
93
|
+
- [Platform Tools](platform-tools/index.mdx#credential-security) - Three-layer credential model, HTTP tool patterns, and credential management
|
|
94
|
+
|
|
95
|
+
### Reference
|
|
96
|
+
|
|
97
|
+
- [Concepts](concepts.mdx) - Plain-English concept explanations, glossary, Zod guide, execution model, and common errors
|
|
98
|
+
- [Templates](templates/index.mdx) - 7 workflow templates: web-scraper, data-enrichment, email-sender, lead-scorer, and more
|
|
99
|
+
- [CLI Reference](cli.mdx) - All CLI commands: check, deploy, exec, resources, executions, env, and more
|
|
100
|
+
- [Deployment](deployment/index.mdx) - Deploy pipeline, versioning, bundle upload, and registry registration
|
|
101
|
+
- [Runtime](runtime.mdx) - Worker execution model, concurrency, timeouts, cancellation, resource limits, and v1 limitations
|
|
102
|
+
|
|
103
|
+
### Framework
|
|
104
|
+
|
|
105
|
+
- [Development Framework](framework/index.mdx) - How Claude Code helps you build: project structure, agent integration, memory, and documentation
|
|
106
|
+
- [Interaction Guidance](framework/interaction-guidance.mdx) - Skill dimension adaptation rules for platform navigation, API integration, and automation concepts
|
|
107
|
+
- [Tutorial System](framework/tutorial-system.mdx) - 21-item tutorial menu, skill-adaptive lesson variants, progress tracking, and module contents
|
|
108
|
+
|
|
109
|
+
### More
|
|
110
|
+
|
|
111
|
+
- [Documentation](framework/index.mdx#resource-documentation) - Writing and deploying MDX documentation alongside your resources
|
|
112
|
+
- [Troubleshooting](troubleshooting.mdx) - Static error catalog for CLI, deployment, schema, and runtime failures
|
|
113
|
+
- [Roadmap](roadmap.mdx) - Planned features including agent execution, streaming logs, and SDK testing utilities
|
|
114
|
+
|
|
115
|
+
---
|
|
116
|
+
|
|
117
|
+
**Last Updated:** 2026-02-25
|