convex-migration-helper

Name: convex-migration-helper
Author: get-convex

byget-convex· 29 skills

Installs

92k

Stars

Forks

How to install convex-migration-helper?

get-convex/agent-skills/convex-migration-helper

$npx -y skills add get-convex/agent-skills --skill convex-migration-helper

Installs into the current project.

›Prefer a prompt? Paste this to your agent

Use this skill

Run `npx skills use "https://github.com/get-convex/agent-skills" --skill "get-convex/agent-skills/convex-migration-helper"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.

Use the whole pack

Use the skills in "https://github.com/get-convex/agent-skills" that are relevant to the current task. Run `npx skills add "https://github.com/get-convex/agent-skills"` and select the relevant skills, then follow their instructions.

Files · 1

View on GitHub

SKILL.md

1	# Convex Migration Helper
2
3	Safely migrate Convex schemas and data when making breaking changes.
4
5	## When to Use
6
7	- Adding new required fields to existing tables
8	- Changing field types or structure
9	- Splitting or merging tables
10	- Renaming or deleting fields
11	- Migrating from nested to relational data
12
13	## When Not to Use
14
15	- Greenfield schema with no existing data in production or dev
16	- Adding optional fields that do not need backfilling
17	- Adding new tables with no existing data to migrate
18	- Adding or removing indexes with no correctness concern
19	- Questions about Convex schema design without a migration need
20
21	## Key Concepts
22
23	### Schema Validation Drives the Workflow
24
25	Convex will not let you deploy a schema that does not match the data at rest.
26	This is the fundamental constraint that shapes every migration:
27
28	- You cannot add a required field if existing documents don't have it
29	- You cannot change a field's type if existing documents have the old type
30	- You cannot remove a field from the schema if existing documents still have it
31
32	This means migrations follow a predictable pattern: **widen the schema, migrate
33	the data, narrow the schema**.
34
35	### Online Migrations
36
37	Convex migrations run online, meaning the app continues serving requests while
38	data is updated asynchronously in batches. During the migration window, your
39	code must handle both old and new data formats.
40
41	### Prefer New Fields Over Changing Types
42
43	When changing the shape of data, create a new field rather than modifying an
44	existing one. This makes the transition safer and easier to roll back.
45
46	### Don't Delete Data
47
48	Unless you are certain, prefer deprecating fields over deleting them. Mark the
49	field as `v.optional` and add a code comment explaining it is deprecated and why
50	it existed.
51
52	## Safe Changes (No Migration Needed)
53
54	### Adding Optional Field
55
56	```typescript
57	// Before
58	users: defineTable({
59	name: v.string(),
60	});
61
62	// After - safe, new field is optional
63	users: defineTable({
64	name: v.string(),
65	bio: v.optional(v.string()),
66	});
67	```
68
69	### Adding New Table
70
71	```typescript
72	posts: defineTable({
73	userId: v.id("users"),
74	title: v.string(),
75	}).index("by_user", ["userId"]);
76	```
77
78	### Adding Index
79
80	```typescript
81	users: defineTable({
82	name: v.string(),
83	email: v.string(),
84	}).index("by_email", ["email"]);
85	```
86
87	## Breaking Changes: The Deployment Workflow
88
89	Every breaking migration follows the same multi-deploy pattern:
90
91	Deploy 1 - Widen the schema:
92
93	1. Update schema to allow both old and new formats (e.g., add optional new
94	field)
95	2. Update code to handle both formats when reading
96	3. Update code to write the new format for new documents
97	4. Deploy
98
99	Between deploys - Migrate data:
100
101	5. Run migration to backfill existing documents
102	6. Verify all documents are migrated
103
104	Deploy 2 - Narrow the schema:
105
106	7. Update schema to require the new format only
107	8. Remove code that handles the old format
108	9. Deploy
109
110	## Using the Migrations Component
111
112	For any non-trivial migration, use the
113	[`@convex-dev/migrations`](https://www.convex.dev/components/migrations)
114	component. It handles batching, cursor-based pagination, state tracking, resume
115	from failure, dry runs, and progress monitoring.
116
117	See `references/migrations-component.md` for installation, setup, defining and
118	running migrations directly with `npx convex run migrations:myMigration`, dry
119	runs, status monitoring, and configuration options.
120
121	## Common Migration Patterns
122
123	See `references/migration-patterns.md` for complete patterns with code examples
124	covering:
125
126	- Adding a required field
127	- Deleting a field
128	- Changing a field type
129	- Splitting nested data into a separate table
130	- Cleaning up orphaned documents
131	- Zero-downtime strategies (dual write, dual read)
132	- Small table shortcut (single internalMutation without the component)
133	- Verifying a migration is complete
134
135	## Common Pitfalls
136
137	1. Making a field required before migrating data: Convex rejects the deploy
138	because existing documents lack the field. Always widen the schema first.
139	2. Using `.collect()` on large tables: Hits transaction limits or causes
140	timeouts. Use the migrations component for proper batched pagination.
141	`.collect()` is only safe for tables you know are small.
142	3. Not writing the new format before migrating: Documents created during the
143	migration window will be missed, leaving unmigrated data after the migration
144	"completes."
145	4. Skipping the dry run: Use `dryRun: true` to validate migration logic
146	before committing changes to production data. Catches bugs before they touch
147	real documents.
148	5. Deleting fields prematurely: Prefer deprecating with `v.optional` and a
149	comment. Only delete after you are confident the data is no longer needed and
150	no code references it.
151	6. Using crons for migration batches: The migrations component handles
152	batching via recursive scheduling internally. Crons require manual cleanup
153	and an extra deploy to remove.
154
155	## Migration Checklist
156
157	- [ ] Identify the breaking change and plan the multi-deploy workflow
158	- [ ] Update schema to allow both old and new formats
159	- [ ] Update code to handle both formats when reading
160	- [ ] Update code to write the new format for new documents
161	- [ ] Deploy widened schema and updated code
162	- [ ] Define migration using the `@convex-dev/migrations` component
163	- [ ] Test with `npx convex run migrations:myMigration '{"dryRun": true}'`
164	- [ ] Run migration directly with `npx convex run migrations:myMigration` and
165	monitor status
166	- [ ] Verify all documents are migrated
167	- [ ] Update schema to require new format only
168	- [ ] Clean up code that handled old format
169	- [ ] Deploy final schema and code
170	- [ ] Remove migration code once confirmed stable

6 picks

Type

Run `npx skills use "https://github.com/get-convex/agent-skills" --skill "get-convex/agent-skills/convex-migration-helper"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.

Use the skills in "https://github.com/get-convex/agent-skills" that are relevant to the current task. Run `npx skills add "https://github.com/get-convex/agent-skills"` and select the relevant skills, then follow their instructions.

convex-migration-helper

How to install convex-migration-helper?

Related

convex-migration-helper

How to install convex-migration-helper?

Related