Setting the file. One moment.
Skill 26 · Bedrock Migration Assessment
Subchapter 26.5
references/shared/schema-preferences.mdMarkdown8 KBView on GitHub
Clarify phase output (references/phases/clarify/clarify-ai-only.md). Design, Estimate, and Generate read it once those phases exist in this skill.
Scripts
Validate AI ProfileEvery key in design_constraints, ai_constraints, and startup_constraints (when present) MUST be an object with:
| Field | Type | Required | Description |
|---|---|---|---|
value | any | yes | Interpreted constraint value |
chosen_by | string | yes | "user" | "default" | "extracted" | "derived" |
prompt | string | yes | Question shown to the user, or a short detection label when skipped |
design_consequence | string | yes | One sentence: how this choice shaped the AWS design, estimate, or plan |
question_id | string | no | AI-only catalog ID (Q1–Q11, Q1.5) when mapped; omit for derived values |
source | string | no | Raw provenance signal. Required when chosen_by is "extracted" (prefix ai-profile:) or "default" ("default:<Qid>"). Omit for "user" and "derived". |
Do not write null values. Omit keys that produce no constraint.
chosen_by | prompt content |
|---|---|
user | Verbatim question text from clarify-ai-only.md (blockquote body, without answer options) |
extracted | Detection label, e.g. "Detected: gpt-4o (openai via openai_api, high confidence)" |
default | Question text + " (default applied)" |
derived | The specific derivation rule, e.g. "Derived from compliance: gdpr → eu-west-1" |
Use the Impact row for the selected answer from clarify-ai-only.md when the user answered. For extracted/default/derived, use the catalog below or synthesize one sentence tied to the actual value.
{
"metadata": {
"migration_type": "ai-only",
"clarify_mode": "full",
"timestamp": "<ISO 8601>",
"discovery_artifacts": ["ai-workload-profile.json"],
"questions_asked": ["Q2"],
"questions_defaulted": ["Q3"],
"questions_extracted": ["Q5"],
"report_caveats": ["Compliance requirements were not confirmed by the user"]
},
"design_constraints": {
"<key>": {
"value": "<answer or default>",
"chosen_by": "user|default|extracted|derived",
"prompt": "<question text shown>",
"design_consequence": "<one line: what this decides>",
"question_id": "<Qn — optional>",
"source": "<provenance — optional>"
}
},
"ai_constraints": {
"<key>": {
"value": "<answer or default>",
"chosen_by": "user|default|extracted|derived",
"prompt": "<question text shown>",
"design_consequence": "<one line: what this decides>",
"question_id": "<Qn — optional>",
"source": "<provenance — optional>"
}
},
"startup_constraints": {
"<key>": {
"value": "<answer or default>",
"chosen_by": "user|default|extracted|derived",
"prompt": "<question text shown>",
"design_consequence": "<one line: what this decides>",
"question_id": "<Qn — optional>",
"source": "<provenance — optional>"
}
}
}metadata.report_caveats is present only when a caveat was added (Q1.5 default).
| Key | Section | question_id | Default prompt (user-asked) | design_consequence template |
|---|---|---|---|---|
target_region | design_constraints | — | Derived from compliance and cross-cloud answers | Bedrock resources deploy in [value]; model availability is checked for this region |
compliance | design_constraints | Q1.5 | Do you have any compliance or regulatory requirements? | [value] gates Bedrock models, regions, and logging. ["unknown"] requires a report caveat |
ai_framework | ai_constraints | Q1 | Which AI framework or orchestration layer are you using? | [value] determines migration effort (gateway config vs SDK swap vs agent framework migration) |
ai_priority | ai_constraints | Q2 | What matters most for your AI application? | [value] drives Bedrock model selection (quality vs cost vs latency) |
ai_monthly_spend | ai_constraints | Q3 | Monthly AI spend on OpenAI or Gemini? | [value] band sets token volume and model tier assumptions |
cross_cloud | ai_constraints | Q4 | Cross-cloud API call concerns | [value] selects endpoint type (public, VPC endpoint, PrivateLink) and rollout approach |
ai_model_baseline | ai_constraints | Q5 | Current model in use? | [value] is the quality/latency baseline for Bedrock comparison |
ai_vision | ai_constraints | Q6 | What input types must the model accept? | [value] requires vision- or audio-capable Bedrock models when not text-only |
ai_token_volume | ai_constraints | Q7 | Monthly AI usage volume | [value] sets on-demand vs provisioned throughput and caching analysis |
ai_latency | ai_constraints | Q8 | Response speed importance | [value] adds latency success criteria and model filtering |
ai_complexity | ai_constraints | Q9 | AI task complexity | [value] affects recommended model size and agentic path |
ai_critical_feature | ai_constraints | Q10 | Specialized features needed | [value] gates the model shortlist and capability validation |
startup_program_status | startup_constraints | Q11 | Have you applied for AWS Activate credits? | [value] triggers the Activate credits callout in the plan |
Before marking Clarify complete:
value, chosen_by, prompt, and design_consequence.prompt or design_consequence.question_id present when the constraint maps to a catalog question.source present on every constraint where chosen_by is "extracted" or "default". Omit for "user" and "derived".