Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add angad-kandhari/deliberate --skill migrategit clone --depth 1 https://github.com/angad-kandhari/deliberateWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/angad-kandhari/deliberate/migrate)<a href="https://agentmods.dev/skills/angad-kandhari/deliberate/migrate"><img src="https://agentmods.dev/badge/skills/angad-kandhari/deliberate/migrate.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00074 | $0.01580 |
| Opus 5 | $0.00037 | $0.00790 |
| Sonnet 5 | $0.00015 | $0.00316 |
| Haiku 4.5 | $0.00007 | $0.00158 |
Grade A, and why
migrate scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Migrate
Migration skill for LLM coding agents. Load this for non-trivial migrations: database schema changes, framework upgrades, API version bumps, rename rollouts, data backfills, or any change that can't land in a single atomic deploy.
Where deliberate keeps code honest, this skill keeps transitions honest. Counters the big-bang rewrite, the half-finished migration, and the "we'll clean up the old path later" that never happens.
1. Incremental By Default
Big-bang migrations fail. Plan for coexistence.
The question isn't "how do I switch from old to new?" It's "how do I run old and new side by side long enough to verify, then cut over?"
Default migration shape:
- Add the new path alongside the old. Don't remove anything yet.
- Dual-write / dual-read where relevant. Old and new stay in sync.
- Migrate callers one at a time. Verify each.
- Backfill any historical data.
- Cut over reads to the new path.
- Remove the old path only after the new path has been stable in production.
A migration that requires all callers to update in the same PR is not a migration. It's a rewrite, and it will stall.
Test: Can I deploy each step independently, and roll any single step back?
2. Write the Rollback Before You Write the Migration
If you can't undo it, you can't ship it.
For every migration step, name the rollback:
- Schema add: drop the column. Safe if nothing writes to it yet.
- Schema drop: restore from backup. Often not safe - push drops to the end.
- Data backfill: idempotent, so reruns don't double-count. Or: clearly one-way, with a rollback plan documented.
- Code cutover: feature flag, so reverting is a config change, not a deploy.
If the rollback is "restore the DB," your migration is too big. Break it up.
Test: If this breaks in production at 3am, how do I undo it without paging the DBA?
3. Make Schema Changes Safe, Not Fast
Every destructive schema change is a minefield.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 169 lines · 74 tokens per session scan A 8ebb63668555
migrate is a skill published in the GitHub repository angad-kandhari/deliberate (2 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 74 tokens to every session and 1,580 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
nestjs-caching-queues
NestJS caching and background-job queue best practices. Use when adding caching or async job processing to NestJS — CacheModule, cache-manager, Redis, or BullMQ/Bull queues. Triggers on CacheModule, @InjectQueue, @Processor, cache TTL, or background jobs.
nestjs-database
NestJS database and ORM best practices (TypeORM, Prisma, Mongoose). Use when integrating a database in NestJS — repositories, entities, migrations, transactions, or query performance. Triggers on TypeOrmModule, @InjectRepository, PrismaService, @Entity, migrations, or N+1 queries.
ecto-patterns
Ecto patterns — schemas, changesets, queries, migrations, Multi, associations, preloads, upserts. Use when editing Repo calls, Ecto.Query, or schema fields. Skip for Ash.
ggsql
Write ggsql queries — a grammar of graphics for SQL. Use when the user wants to create, modify, or understand a ggsql visualization query.
bun-redis
Use when working with Redis in Bun (ioredis, Upstash), caching, pub/sub, session storage, or key-value operations.
bun-sqlite
Use for bun:sqlite, SQLite operations, prepared statements, transactions, and queries.