> For the complete documentation index, see [llms.txt](https://experience-digital-transform.gitbook.io/experience-digital-transformation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://experience-digital-transform.gitbook.io/experience-digital-transformation/cloud-for-dpi/agent-skills/learning-system/cloud-decision-gap-discovery.md). # Cloud Decision Gap Discovery Surface unanswered questions and evidence needs. {% hint style="info" %} **Skill** `cloud-decision-gap-discovery` · version 0.4.0 · [Chapter 1](/experience-digital-transformation/cloud-for-dpi/chapter-1.md) · [Source](https://git.vanillacore.net/giga/cloud-skills/src/commit/5ef20352d7e160cb5145c3773a7f002e1f03d5a3/skills/cloud-decision-gap-discovery) · [Install and use](/experience-digital-transformation/cloud-for-dpi/agent-skills/install-and-use.md) {% endhint %} Use this skill when a cloud or hosting question is premature, contradictory, or missing local evidence. It turns uncertainty into bounded evidence work rather than plausible assumptions. ## When to use * Required service, architecture, contract, operating, recovery, or exit evidence is absent. * Specialist outputs disagree. * A learner needs to decide whether to continue, investigate, pilot, or pause. Do not use it to turn uncertainty into a recommendation or to assign an owner that was not supplied. ## Input ```yaml mode: ask | learn | work | challenge | review case_id: decision_under_review: upstream_artifacts: [] supplied_evidence: [] learner_answer: ``` ## Procedure 1. Restate the exact decision or continuation gate. 2. Collect referenced `fact_id`, `finding_id`, and `gap_id` values from upstream artifacts. Do not replace identifiers with loose prose. 3. Separate questions answerable from named public sources from questions requiring local legal, contractual, security, operating, capacity, routing, recovery, or exit evidence. 4. Classify each gap as `decision_blocking`, `before_commitment`, `reversible_pilot`, or `monitor`. 5. Record the affected decision, evidence needed, validation method, evidence owner or `local_owner_required`, consequence, and permitted next route. 6. If foundational service facts are absent, stop option comparison. If specialist findings conflict, create a reconciliation gap rather than choosing one silently. 7. Return `decision_gaps/v1` and one bounded action. ## Learner modes In `challenge` mode, if no learner claim or attempt is supplied, pose one bounded challenge and stop before the model correction. After the attempt, identify the first unsupported step, explain the correction, and give one retry. * **Ask:** Explain why the named gap blocks or qualifies the decision. * **Learn:** Present facts and assumptions together and ask the learner to identify which require local evidence. * **Work:** Produce the evidence register from the supplied artifacts. * **Challenge:** Ask the learner which conclusion fails if the evidence is absent before revealing the classification. * **Review:** Check traceability, priority, validation feasibility, ownership, and permitted next route. ## Output ```yaml decision_gaps: schema: decision-gaps/v1 case_id: gaps: - gap_id: blocked_finding_ids: [] question: decision_impact: priority: decision_blocking | before_commitment | reversible_pilot | monitor evidence_needed: validation_method: evidence_owner: supplied_role | local_owner_required consequence_if_unresolved: next_route: ``` ## Challenge checks Reject a gap record that uses “confirm compliance,” “test resilience,” or similar language without naming the object, method, success condition, owner, and decision consequence. Reject dates, owners, environment values, or successful results that were not supplied. ## Sources Use the applicable immutable chapter sources listed in the installed sibling skill or the repository source map. Local evidence remains a requested input, not a public citation. If no source map is available, ask for the controlling public revision rather than inventing one. ## Pitfalls * A long question list is not prioritisation. * `monitor` is not appropriate for evidence required before an irreversible commitment. * A reversible pilot needs a bounded scope, success and stop conditions, cleanup, and an owner. ## Verification The result passes when every gap is traceable, actionable, correctly prioritised, assigned only to a supplied or explicitly unresolved owner, and connected to a next route that does not outrun the evidence. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://experience-digital-transform.gitbook.io/experience-digital-transformation/cloud-for-dpi/agent-skills/learning-system/cloud-decision-gap-discovery.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.