spec-migration

spec-migration is a skill for Claude Code, Codex from Chemaclass/agnostic-ai. It costs 68 tokens per session (1,304 once invoked), scanned A, original, MIT.

A guide for deciding when changes to configuration specifications need an automatic migration. A migration rewrites older configuration into a newer form so users do not have to edit it by hand.

In plain words
What is it for?
Use it when changing fields in files under .agnostic-ai or in agnostic-ai.yaml, to decide what migration and warning work to add.
Why use it?
It helps prevent upgrades from breaking specifications that previously loaded, while distinguishing migrations from warnings and behavior changes.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it when changing fields in files under .agnostic-ai or in agnostic-ai.yaml, to decide what migration and warning work to add.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/chemaclass/agnostic-ai/spec-migration
Install

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.

Any agent
npx skills add Chemaclass/agnostic-ai --skill spec-migration
Clone the repo
git clone --depth 1 https://github.com/Chemaclass/agnostic-ai

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for spec-migration

README.md
[![agentmods](https://agentmods.dev/badge/skills/chemaclass/agnostic-ai/spec-migration/github.svg)](https://agentmods.dev/skills/chemaclass/agnostic-ai/spec-migration)
Your own site
<a href="https://agentmods.dev/skills/chemaclass/agnostic-ai/spec-migration"><img src="https://agentmods.dev/badge/skills/chemaclass/agnostic-ai/spec-migration/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for spec-migration

Your own site · 80×15
<a href="https://agentmods.dev/skills/chemaclass/agnostic-ai/spec-migration"><img src="https://agentmods.dev/badge/skills/chemaclass/agnostic-ai/spec-migration.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 68 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,304 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00068 $0.01304
Opus 5.5 $0.00027 $0.00522
Sonnet 5.5 $0.00014 $0.00261
Haiku 4.5 $0.00007 $0.00130

Measured 6d ago against content hash 53faaaed152a, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

spec-migration 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 6d 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.

.agnostic-ai/skills/spec-migration/SKILL.md · 51 lines

How it starts

The opening of the file, as written. The whole thing — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.

spec-migration

A spec migration rewrites a user's old spec form into the new one without changing what sync writes. agnostic-ai migrate runs every pending migration from one registry, specMigrations in internal/cli/migrate.go (#1755). Users should never have to rewrite specs by hand after an upgrade.

When a change needs one

Ask this for every change to a spec kind, a frontmatter field, a YAML key, or agnostic-ai.yaml:

The change What to ship
Renames a field, or adds a new form that some or all entries map to one to one A migration and a lint warning on the old form; the old form stays accepted
Deprecates a form that no entry maps to one to one No migration: a lint warning that explains the manual rewrite
Adds a new optional field Nothing
Rejects specs that used to load, where a rewrite that keeps their meaning makes them load again A migration, plus a lint warning at least one release before the rejection
Changes a form's meaning or default Not a migration: a behavior change with a CHANGELOG entry and a lint warning at least one release ahead
Removes a form Only in a breaking release, after its migration shipped, no earlier than the migration's issue allows
A style preference with no change in meaning or output Nothing, or a lint note. Never a migration

While the project is 0.x, a breaking release is a minor release whose CHANGELOG has a breaking section.

Rules

  1. Output stays the same. sync, then migrate, then sync --check exits 0 for every target the spec already reached. The only exceptions are the ones the migration's issue names, such as a spec reaching new targets (which sync then lists) or a literal secret becoming a reference. Any other change to a synced file is not a migration.
  2. Old form keeps working until its removal release. That release replaces it with an error that names the migration to run with the last version that has it. The migration and its fixture stay until then.
  3. Safe to run twice, and stateless. A migration detects its own old form. Running it twice changes nothing. The release it records is metadata only.
  4. Map one to one or skip. Entries that do not map stay as written with a one-line reason that migrate reports. A spec that sets both the old and the new form is a skip, never a merge.
  5. Touch only what you own. Edit through the shared editor in internal/cli/migrate_yaml.go: rewriteTopLevelYAMLKeys for a YAML spec, rewriteFrontmatterKeys for Markdown frontmatter. Comments, key order, quoting, and bodies stay as written. The registry writes atomically and keeps the file mode.
  6. Never expose a secret. Diffs, skip reasons, and errors show values of env, headers, URLs, and args as references or <redacted>; redactMigrationLines does this for every migration. A migration never turns a reference into a literal and never moves a value into another file or into the global home.
  7. Stay inside the spec roots. The registry resolves each change's real path. A file outside the project, local/, or global spec roots, such as a pack or a symlink into one, becomes a skip that names the pack. Skip a pack layer's entry in Plan with packSkip.
  8. Import writes the new form. A project that starts after the change never needs the migration.

Read the full file on GitHub · 51 lines

Changes

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.

  1. 6d ago Changed 53faaaed152a
  2. 11d ago First seen · 51 lines · 68 tokens per session scan A 1c9a8d2d0771

Subscribe to this mod's changes

spec-migration is a skill published in the GitHub repository Chemaclass/agnostic-ai (25 stars, last pushed today), licensed MIT. It adds 68 tokens to every session and 1,304 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-05.