Skip to content
Development
Skill

/setup-validate

Validate a `setup.aiwg.io/v1` SetupManifest file against the schema and run cons

From plugin
aiwg
229200 skills199 agents26 commands

Claude Code

$ npx -y skills add jmagly/aiwg --skill setup-validate --agent claude-code

SKILL.md

setup-validate.SKILL.md
namespace: aiwg
name: setup-validate
platforms: [all]
description: "Validate a `setup.aiwg.io/v1` SetupManifest file against the schema and run cons"

setup-validate

Validate a `setup.aiwg.io/v1` SetupManifest file against the schema and run consistency checks.

For the public AIWG installer, retain `metadata.execution_mode: provider-orchestrated`. Validate the manifest's interview → review → apply sequence; do not convert it to deterministic execution. Guided plan review (`aiwg setup --review`) is distinct from SetupManifest schema validation.

Trigger Phrases

  • "validate setup manifest"
  • "check setup.manifest.yaml"
  • "aiwg setup-validate [manifest]"
  • "lint installer manifest"
  • "verify my setup manifest"

Parameters

manifest (positional, optional)

Path to the `setup.manifest.yaml`. Default: `./setup.manifest.yaml`.

--schema (optional)

Path to schema file. Default: auto-located from AIWG installation.

--strict (optional)

Fail on warnings in addition to errors.

--fix (optional)

Reserved for future safe autofixes. The current CLI validates and reports only.

Validation Checks

Schema Validation

  • `apiVersion` must be `setup.aiwg.io/v1`
  • `kind` must be `SetupManifest`
  • Required fields present: `metadata.name`, `spec.platforms`, `spec.steps`
  • All step types are one of: `script`, `detect`, `ask`, `verify`, `agentic`, `platform-route`, `chain`, `os-config`
  • Param types are one of: `string`, `path`, `boolean`, `integer`, `choice`
  • `metadata.install_type` must be one of: `user`, `developer`, `ci` (if present)
  • `metadata.execution_mode` must be `deterministic` or `provider-orchestrated` (if present)

Reference Checks

  • Every `script:` path exists relative to the manifest directory
  • Every `chain:` manifest path exists
  • Every `platform-route` route value resolves to an existing script
  • `depends_on` step IDs all exist in the manifest
  • Recovery `triggers` list references valid step IDs
  • `os-config` steps: `config_id` must reference a valid entry in `spec.os_config`

Developer Manifest Checks (when `metadata.install_type: developer`)

These checks apply only to manifests with `install_type: developer`:

Errors

  • `os-config` steps must reference a valid `config_id` in the `spec.os_config` block
  • `interactive_required: true` params must not have a `default` value — a default defeats the purpose of requiring interactive input
  • Each platform declared in `spec.platforms` must have at least one platform-specific step or `os_config` entry

Warnings (unless `--strict`, then errors)

  • Developer manifest with no `os_config` block: warn unless `metadata.description` explicitly states why no OS configuration is needed
  • `os_config` entries with `requires_relogin: true` but no corresponding `os-config` step — the entry is declared but never applied
  • `os_config` entries with `interactive: true` but no user-facing explanation in `description`

Consistency Checks (warnings unless --strict)

  • Steps without `id` — suggest auto-generated IDs
  • Sequential steps that logically depend on prior steps but have no `depends_on`
  • Params declared but never referenced in scripts or `when` conditions
  • Prerequisites with no `install_hint` (hard to debug without one)
  • Verify expressions that look like they'd always pass (e.g., `true`)
  • `interactive_required: true` params that also have `required: true` — the `required` flag is redundant when `interactive_required` is set

Script Safety Checks

  • Scripts that exist: read first 5 lines to confirm they source lib scripts
  • Warn if a script template is used directly without customization (template placeholder text detected)

Agentic Step Audit

  • Warn on any `type: agentic` step in a deterministic manifest — print a reminder that agentic steps are exception handling only
  • Do not warn merely because a provider-orchestrated manifest uses agentic steps
  • Error if `type: agentic` step has no `instruction` field

Output Format

Validating: setup.dev.manifest.yaml

  Schema:        ✓ Valid (setup.aiwg.io/v1 / SetupManifest)
  Install type:  developer
  Metadata:      ✓ name=myapp-dev version=1.0.0
  Platform:      ✓ linux (ubuntu, debian), macos
  Params:        ✓ 4 params (INSTALL_DIR required, LOCAL_DOMAIN interactive_required, SSH_EMAIL interactive_required, IDE interactive_required)
  Prerequisites: ✓ 2 checked (git >=2.30, docker)
  OS Config:     ✓ 3 entries (docker-group [linux, requires_relogin], inotify-watches [linux], xcode-cli-tools [macos, interactive])

  Steps (7):
    ✓ clone             script      installer/scripts/clone.sh [exists]
    ✓ setup-git-config  script      installer/scripts/dev/setup-git-config.sh [exists]
    ✓ setup-ssh-key     script      installer/scripts/dev/setup-ssh-key.sh [exists]
    ✓ apply-docker-group  os-config  config_id=docker-group [valid]
    ✓ apply-inotify     os-config   config_id=inotify-watches [valid]
    ✓ configure-dev     script      installer/scripts/dev/configure-dev.sh [exists]
    ✓ verify-dev        script      installer/scripts/dev/verify-dev.sh [exists]

  Recovery (1):
    ✓ full-reset  script [exists]

  Warnings (1):
    ⚠ xcode-cli-tools os_config entry declared but no os-config step applies it

Result: VALID (1 warning)

Error Example

  Errors (2):
    ✗ Step 'apply-docker-group': config_id 'docker-group' not found in os_config block
    ✗ Param 'LOCAL_DOMAIN': interactive_required=true but has a default value — remove the default

Result: INVALID — fix errors before running

Auto-fix (--fix)

`--fix` is reserved for future safe autofixes. The current agent-facing CLI validates and reports only; it does not mutate manifests.

References

  • Schema: `agentic/code/addons/agentic-installer/schemas/v1/setup-manifest.schema.json`
  • Run skill: `agentic/code/addons/agentic-installer/skills/setup-run/SKILL.md`
  • Generate skill: `agentic/code/addons/agentic-installer/skills/setup-generate/SKILL.md`
Read more
@jmagly
Ships withaiwg

Reusable project context and specialist workflows for the AI tools you already use. Plan software, coordinate specialist reviews, prepare campaigns, investigate incidents, organize research, curate media, and maintain operational knowledge.

229200 skills199 agents
Get the whole plugin

Stars on GitHub

Momentum

stars in total
229
Star history of jmagly/aiwg on GitHub, a chart by star-history.com

Other skills on aiwg.

agent-loop-ext

agent-loop-ext

Crash-resilient external agent loop with state persistence and CI/CD integration

@jmagly@jmaglyInstall Skill
agent-loop

agent-loop

Detect requests for iterative autonomous agent loops and route to the appropriate loop executor

@jmagly@jmaglyInstall Skill
auto-test-execution

auto-test-execution

Automatically execute tests when code-generating agents modify source files, enforcing the execute-before-return pattern

@jmagly@jmaglyInstall Skill

More skills from other plugins.

brainstorming
Auto-invoked

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or…

@obra@obraInstall Skill
ask-matt

ask-matt

Ask which skill or flow fits your situation. A router over the skills in this repo.

skills12d309k
@mattpocock@mattpocockInstall Skill
accessibility
Auto-invoked

accessibility

Design, implement, and audit accessible UI to WCAG 2.2 Level AA across Web, iOS, and Android — semantic ARIA roles and labels,…

ecc13d293k
@affaan-m@affaan-mInstall Skill
compiler-commit

compiler-commit

Use when you want to verify compiler changes and commit with the correct convention. Runs tests, lint, and format, then commits…

react2d252k
@facebook@facebookInstall Skill

Questions about setup-validate.

Which coding agents can use setup-validate?
setup-validate installs in Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode and Windsurf with the skills CLI, using each agent's flag shown on this page. Flowy's auto-invocation, which fires a skill when your prompt matches, runs in Claude Code only.