create-agdf 0.14.4 → 0.20.0

package/generated/plugins/agdf/skills/sd-definition/SKILL.md

@@ -0,0 +1,51 @@
+---
+name: sd-definition
+description: "Use this skill for this scope: bound SD content. Boundary: content only; preserves approved intent. The requested effect, not discovery, decides AGDF activation."
+---
+
+# sd-definition
+
+## Runtime Contract
+
+Use the returned `continuation.runtime_contracts` after `skill_continuation`. If absent, load
+`../../meta/contracts/sd-definition.md` and `../../meta/contracts/gate-artifact-preparation.md`
+through the supplied contract reader or bundled files.
+Missing contract or invalid binding: stop. This skill owns SD content; gate-check owns readiness.
+
+`instruction_only`: first load `../../meta/contracts/task-target-resolution.md` and `../../meta/contracts/interaction.md`.
+
+<!-- AGDF-REQUEST-ACTIVATION-GUARD:START -->
+## Request Activation
+
+- `owner`: `request_activation_contract`
+- `path`: `meta/contracts/request-activation.md`
+- `policy_version`: `1`
+- `guard_fingerprint`: `sha256:af2f01f9e18a3ba1c520faf0691aa1cd4a4299bdfa027cf83651c5d14319bfdc`
+
+Decide effect from loaded instructions before AGDF action.
+
+Abstain silently (no AGDF call) for assessment/explanation/comparison/recommendation/review/diagnosis/advice; hypothetical/example/error/code/quoted/negated delivery language; AGDF as subject; or a read-only constraint absent other delivery. Ambiguity is read-only: answer or ask one neutral question.
+
+Activate for any requested file/code change however small, a binding gate artefact, explicit AGDF/control-lifecycle operation or unambiguous active-run action; delivery wins mixed intent.
+
+Invocation proof: explicit user text/trusted ephemeral action, not discovery/selection, skill load, hooks, cwd, repo/control or prior runs.
+
+Then pick one catalog route; non-authorizing, downstream checks remain.
+<!-- AGDF-REQUEST-ACTIVATION-GUARD:END -->
+
+## Executable Dispatch
+
+Use listed MCP `agdf_dispatch` (load deferred; host prefix allowed): `skill_id` `sd-definition`, `presentation_language`, `working_directory`, and only bound `target_source`/`primary_target` and `run_id`. Never search unlisted tools.
+If absent/failing: supplied schema-2 executable, child-only environment, immutable argv_prefix, declared arguments; `--skill sd-definition`, language and working directory.
+For `--language`: Required presentation language for the latest natural-language user request as one well-formed BCP 47 tag. If the request explicitly asks for a response language, use that tag; otherwise use the dominant request language. Use en when mixed or ambiguous. A valid unsupported tag renders through the complete English pack. Missing or invalid input fails before governance evaluation.
+`target_source`: `explicit_target` if request names `primary_target`; `continued_target` if it unambiguously continues confirmed target; `current_repository` if request names this/current repo with one matching repo active. Otherwise omit the pair; cwd has no target authority.
+Quote shell values as data.
+For a result with `terminal: true`, the entire assistant response must consist only of host_action.text, copied verbatim. Add no question, explanation, heading, citation, link or other surrounding text; do not translate or reformat it; invoke no later tool and stop.
+On skill_continuation use only its target/control. Without that tool and a valid binding: `dispatcher_unavailable`; no runtime search, environment repair
+or help retries. Dispatch never authorizes.
+
+## Workflow
+
+Follow the focused SD contract using the original request and answered conversation context.
+Write only the supplied unapproved SD from approved PRD and analyses, preserving uncertainty and explicit human approval.
+After canonical recording, redispatch gate-check; use its fresh presentation and wait.

package/generated/plugins/agdf/skills/sd-definition/help.md

@@ -0,0 +1,17 @@
+## NAME
+
+`sd-definition` - draft and clarify a bound unapproved solution design
+
+## SYNOPSIS
+
+`/sd-definition` [`--help`|`-h`]
+
+## DESCRIPTION
+
+Derives technical ownership, solution decisions and acceptance traceability from the approved
+PRD and applicable analyses after bound dispatch. Keeps material unresolved decisions visible.
+Draft revision uses canonical recording and a fresh presentation before human approval.
+
+## SEE ALSO
+
+See SKILL.md and the focused SD contract for binding, intent, revision and approval boundaries.

package/generated/plugins/agdf/skills/task-plan-review/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: task-plan-review
-description: Use this skill after code changes and before QA to verify whether the approved Task Plan was actually fulfilled. It checks each task_id against implementation, acceptance criteria, tests, visible evidence, and deviations, and provides TP coverage for QA.
+description: "Use this skill for this scope: evidence dimension: verify whether the approved Task Plan was fulfilled. Boundary: supports Quality Readiness; no final QA decision. The requested effect, not discovery, decides AGDF activation."
 ---
 
 # task-plan-review
@@ -19,12 +19,44 @@ For each relevant task, this skill determines:
 This skill provides TP coverage. It is not the final QA decision.
 
 ## Runtime Contract
-Use these focused runtime-contract modules:
+Use `continuation.runtime_contracts` after `skill_continuation`. If absent: listed MCP `agdf_inspect` (`operation: contract`, `module: <name>`); else supplied schema-2 executable/argv_prefix[0] `contract --module <name>`; else bundled files below. MCP reads need no hook or shell. Missing module: stop; never infer/search for runtime.
 
 - `../../meta/contracts/quality.md`
 - `../../meta/contracts/context-graph.md`
 - `../../meta/contracts/gate-transition.md`
 
+`instruction_only`: first load `../../meta/contracts/task-target-resolution.md` and `../../meta/contracts/interaction.md`.
+
+<!-- AGDF-REQUEST-ACTIVATION-GUARD:START -->
+## Request Activation
+
+- `owner`: `request_activation_contract`
+- `path`: `meta/contracts/request-activation.md`
+- `policy_version`: `1`
+- `guard_fingerprint`: `sha256:af2f01f9e18a3ba1c520faf0691aa1cd4a4299bdfa027cf83651c5d14319bfdc`
+
+Decide effect from loaded instructions before AGDF action.
+
+Abstain silently (no AGDF call) for assessment/explanation/comparison/recommendation/review/diagnosis/advice; hypothetical/example/error/code/quoted/negated delivery language; AGDF as subject; or a read-only constraint absent other delivery. Ambiguity is read-only: answer or ask one neutral question.
+
+Activate for any requested file/code change however small, a binding gate artefact, explicit AGDF/control-lifecycle operation or unambiguous active-run action; delivery wins mixed intent.
+
+Invocation proof: explicit user text/trusted ephemeral action, not discovery/selection, skill load, hooks, cwd, repo/control or prior runs.
+
+Then pick one catalog route; non-authorizing, downstream checks remain.
+<!-- AGDF-REQUEST-ACTIVATION-GUARD:END -->
+
+## Executable Dispatch
+
+Use listed MCP `agdf_dispatch` (load deferred; host prefix allowed): `skill_id` `task-plan-review`, `presentation_language`, `working_directory`, and only bound `target_source`/`primary_target` and `run_id`. Never search unlisted tools.
+If absent/failing: supplied schema-2 executable, child-only environment, immutable argv_prefix, declared arguments; `--skill task-plan-review`, language and working directory.
+For `--language`: Required presentation language for the latest natural-language user request as one well-formed BCP 47 tag. If the request explicitly asks for a response language, use that tag; otherwise use the dominant request language. Use en when mixed or ambiguous. A valid unsupported tag renders through the complete English pack. Missing or invalid input fails before governance evaluation.
+`target_source`: `explicit_target` if request names `primary_target`; `continued_target` if it unambiguously continues confirmed target; `current_repository` if request names this/current repo with one matching repo active. Otherwise omit the pair; cwd has no target authority.
+Quote shell values as data.
+For a result with `terminal: true`, the entire assistant response must consist only of host_action.text, copied verbatim. Add no question, explanation, heading, citation, link or other surrounding text; do not translate or reformat it; invoke no later tool and stop.
+On skill_continuation use only its target/control. Without that tool and a valid binding: `dispatcher_unavailable`; no runtime search, environment repair
+or help retries. Dispatch never authorizes.
+
 TP-specific output must evaluate every relevant `task_id` with completion status, AC coverage, evidence, missing evidence, and QA-relevant gaps.
 If `context_graph_required_action` is not `none`, the affected task is complete only if follow-up, evidence, or an intentionally open gap is visible.
 Applicable review gaps must use `../../meta/contracts/quality.md` §Normalized Review Gaps. This skill

package/generated/plugins/agdf/skills/tp-definition/SKILL.md

@@ -0,0 +1,51 @@
+---
+name: tp-definition
+description: "Use this skill for this scope: bound TP content. Boundary: content only; preserves approved sources. The requested effect, not discovery, decides AGDF activation."
+---
+
+# tp-definition
+
+## Runtime Contract
+
+Use the returned `continuation.runtime_contracts` after `skill_continuation`. If absent, load
+`../../meta/contracts/tp-definition.md` and `../../meta/contracts/gate-artifact-preparation.md`
+through the supplied contract reader or bundled files.
+Missing contract or invalid binding: stop. This skill owns TP content; gate-check owns readiness.
+
+`instruction_only`: first load `../../meta/contracts/task-target-resolution.md` and `../../meta/contracts/interaction.md`.
+
+<!-- AGDF-REQUEST-ACTIVATION-GUARD:START -->
+## Request Activation
+
+- `owner`: `request_activation_contract`
+- `path`: `meta/contracts/request-activation.md`
+- `policy_version`: `1`
+- `guard_fingerprint`: `sha256:af2f01f9e18a3ba1c520faf0691aa1cd4a4299bdfa027cf83651c5d14319bfdc`
+
+Decide effect from loaded instructions before AGDF action.
+
+Abstain silently (no AGDF call) for assessment/explanation/comparison/recommendation/review/diagnosis/advice; hypothetical/example/error/code/quoted/negated delivery language; AGDF as subject; or a read-only constraint absent other delivery. Ambiguity is read-only: answer or ask one neutral question.
+
+Activate for any requested file/code change however small, a binding gate artefact, explicit AGDF/control-lifecycle operation or unambiguous active-run action; delivery wins mixed intent.
+
+Invocation proof: explicit user text/trusted ephemeral action, not discovery/selection, skill load, hooks, cwd, repo/control or prior runs.
+
+Then pick one catalog route; non-authorizing, downstream checks remain.
+<!-- AGDF-REQUEST-ACTIVATION-GUARD:END -->
+
+## Executable Dispatch
+
+Use listed MCP `agdf_dispatch` (load deferred; host prefix allowed): `skill_id` `tp-definition`, `presentation_language`, `working_directory`, and only bound `target_source`/`primary_target` and `run_id`. Never search unlisted tools.
+If absent/failing: supplied schema-2 executable, child-only environment, immutable argv_prefix, declared arguments; `--skill tp-definition`, language and working directory.
+For `--language`: Required presentation language for the latest natural-language user request as one well-formed BCP 47 tag. If the request explicitly asks for a response language, use that tag; otherwise use the dominant request language. Use en when mixed or ambiguous. A valid unsupported tag renders through the complete English pack. Missing or invalid input fails before governance evaluation.
+`target_source`: `explicit_target` if request names `primary_target`; `continued_target` if it unambiguously continues confirmed target; `current_repository` if request names this/current repo with one matching repo active. Otherwise omit the pair; cwd has no target authority.
+Quote shell values as data.
+For a result with `terminal: true`, the entire assistant response must consist only of host_action.text, copied verbatim. Add no question, explanation, heading, citation, link or other surrounding text; do not translate or reformat it; invoke no later tool and stop.
+On skill_continuation use only its target/control. Without that tool and a valid binding: `dispatcher_unavailable`; no runtime search, environment repair
+or help retries. Dispatch never authorizes.
+
+## Workflow
+
+Follow the focused TP contract using the original request and answered conversation context.
+Write only the supplied unapproved TP from approved PRD/SD and analyses, preserving uncertainty and explicit human approval.
+After canonical recording, redispatch gate-check; use its fresh presentation and wait.

package/generated/plugins/agdf/skills/tp-definition/help.md

@@ -0,0 +1,17 @@
+## NAME
+
+`tp-definition` - draft and clarify a bound unapproved task/test plan
+
+## SYNOPSIS
+
+`/tp-definition` [`--help`|`-h`]
+
+## DESCRIPTION
+
+Derive tasks, scenarios and executable evidence obligations from approved PRD/SD.
+Block material questions and route concrete clarification to the existing source owner.
+Use existing revision, recording and fresh deliberate TP approval.
+
+## SEE ALSO
+
+See SKILL.md and the focused TP contract.

package/generated/plugins/agdf/skills/ur-definition/SKILL.md

@@ -0,0 +1,54 @@
+---
+name: ur-definition
+description: "Use this skill for this scope: bound UR drafting, clarification or revision. Boundary: content only; preserves approved intent. The requested effect, not discovery, decides AGDF activation."
+---
+
+# ur-definition
+
+## Runtime Contract
+
+Use the returned `continuation.runtime_contracts` after `skill_continuation`. If absent, load
+`../../meta/contracts/ur-definition.md` through the supplied contract reader or bundled file.
+Missing contract or invalid binding: stop. This skill owns UR content; gate-check owns readiness.
+
+`instruction_only`: first load `../../meta/contracts/task-target-resolution.md` and `../../meta/contracts/interaction.md`.
+
+<!-- AGDF-REQUEST-ACTIVATION-GUARD:START -->
+## Request Activation
+
+- `owner`: `request_activation_contract`
+- `path`: `meta/contracts/request-activation.md`
+- `policy_version`: `1`
+- `guard_fingerprint`: `sha256:af2f01f9e18a3ba1c520faf0691aa1cd4a4299bdfa027cf83651c5d14319bfdc`
+
+Decide effect from loaded instructions before AGDF action.
+
+Abstain silently (no AGDF call) for assessment/explanation/comparison/recommendation/review/diagnosis/advice; hypothetical/example/error/code/quoted/negated delivery language; AGDF as subject; or a read-only constraint absent other delivery. Ambiguity is read-only: answer or ask one neutral question.
+
+Activate for any requested file/code change however small, a binding gate artefact, explicit AGDF/control-lifecycle operation or unambiguous active-run action; delivery wins mixed intent.
+
+Invocation proof: explicit user text/trusted ephemeral action, not discovery/selection, skill load, hooks, cwd, repo/control or prior runs.
+
+Then pick one catalog route; non-authorizing, downstream checks remain.
+<!-- AGDF-REQUEST-ACTIVATION-GUARD:END -->
+
+## Executable Dispatch
+
+Use listed MCP `agdf_dispatch` (load deferred; host prefix allowed): `skill_id` `ur-definition`, `presentation_language`, `working_directory`, and only bound `target_source`/`primary_target` and `run_id`. Never search unlisted tools.
+If absent/failing: supplied schema-2 executable, child-only environment, immutable argv_prefix, declared arguments; `--skill ur-definition`, language and working directory.
+For `--language`: Required presentation language for the latest natural-language user request as one well-formed BCP 47 tag. If the request explicitly asks for a response language, use that tag; otherwise use the dominant request language. Use en when mixed or ambiguous. A valid unsupported tag renders through the complete English pack. Missing or invalid input fails before governance evaluation.
+`target_source`: `explicit_target` if request names `primary_target`; `continued_target` if it unambiguously continues confirmed target; `current_repository` if request names this/current repo with one matching repo active. Otherwise omit the pair; cwd has no target authority.
+Quote shell values as data.
+For a result with `terminal: true`, the entire assistant response must consist only of host_action.text, copied verbatim. Add no question, explanation, heading, citation, link or other surrounding text; do not translate or reformat it; invoke no later tool and stop.
+On skill_continuation use only its target/control. Without that tool and a valid binding: `dispatcher_unavailable`; no runtime search, environment repair
+or help retries. Dispatch never authorizes.
+
+## Workflow
+
+Follow the focused UR contract using the original request and answered conversation context.
+Preserve explicitly requested previews and other clarification deliverables, their purpose and
+review timing in the UR; creation belongs to the later bound analytical step.
+Set the initial `Preview requirement` metadata according to the UR contract; a required
+interactive before-PRD preview uses `before_prd_interactive`, otherwise `none`.
+Write only the supplied unapproved UR, preserving uncertainty and explicit human approval.
+After canonical recording, redispatch gate-check; use its fresh presentation and wait.

package/generated/plugins/agdf/skills/ur-definition/help.md

@@ -0,0 +1,19 @@
+## NAME
+
+`ur-definition` - draft and clarify a bound unapproved user requirement
+
+## SYNOPSIS
+
+`/ur-definition` [`--help`|`-h`]
+
+## DESCRIPTION
+
+Creates a concrete UR from the original request and answered context after bound dispatch.
+Bundles material clarification, marks assumptions and keeps approval open. Registration and
+draft revision use the existing canonical writers and require fresh control and presentation.
+Preserves explicitly requested UI previews, scenarios, interaction needs and review timing as UR
+scope for later UX/PRD clarification; it does not create the preview during UR authoring.
+
+## SEE ALSO
+
+See SKILL.md and the focused UR contract for binding, revision and approval boundaries.

package/generated/plugins/agdf/skills/ux-intent-definition/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: ux-intent-definition
-description: Use this skill after approved UR and post-UR routing when UI/UX impact is medium or high, or low with ambiguous product semantics, to define user intent, success, working modes, effective state, visible state ownership, activation, blockers, recovery and transitions as structured PRD input; fail closed when evidence or product decisions are unresolved. This internal analysis never grants gate permission or replaces the approved PRD.
+description: "Use this skill for this scope: approved UR with routed medium/high UX impact or ambiguous low-impact semantics before PRD. Boundary: analytical PRD input only; no intent, gate, design or parallel product authority. The requested effect, not discovery, decides AGDF activation."
 ---
 
 # ux-intent-definition
@@ -12,11 +12,43 @@ approved product promise instead of inventing requirements after code exists.
 
 ## Runtime Contract
 
-Use:
+Use `continuation.runtime_contracts` after `skill_continuation`. If absent: listed MCP `agdf_inspect` (`operation: contract`, `module: <name>`); else supplied schema-2 executable/argv_prefix[0] `contract --module <name>`; else bundled files below. MCP reads need no hook or shell. Missing module: stop; never infer/search for runtime.
 
 - `../../meta/contracts/gate-transition.md` for impact routing, authority and revision behavior;
 - `../../meta/contracts/quality.md` for evidence and fail-closed output discipline.
 
+`instruction_only`: first load `../../meta/contracts/task-target-resolution.md` and `../../meta/contracts/interaction.md`.
+
+<!-- AGDF-REQUEST-ACTIVATION-GUARD:START -->
+## Request Activation
+
+- `owner`: `request_activation_contract`
+- `path`: `meta/contracts/request-activation.md`
+- `policy_version`: `1`
+- `guard_fingerprint`: `sha256:af2f01f9e18a3ba1c520faf0691aa1cd4a4299bdfa027cf83651c5d14319bfdc`
+
+Decide effect from loaded instructions before AGDF action.
+
+Abstain silently (no AGDF call) for assessment/explanation/comparison/recommendation/review/diagnosis/advice; hypothetical/example/error/code/quoted/negated delivery language; AGDF as subject; or a read-only constraint absent other delivery. Ambiguity is read-only: answer or ask one neutral question.
+
+Activate for any requested file/code change however small, a binding gate artefact, explicit AGDF/control-lifecycle operation or unambiguous active-run action; delivery wins mixed intent.
+
+Invocation proof: explicit user text/trusted ephemeral action, not discovery/selection, skill load, hooks, cwd, repo/control or prior runs.
+
+Then pick one catalog route; non-authorizing, downstream checks remain.
+<!-- AGDF-REQUEST-ACTIVATION-GUARD:END -->
+
+## Executable Dispatch
+
+Use listed MCP `agdf_dispatch` (load deferred; host prefix allowed): `skill_id` `ux-intent-definition`, `presentation_language`, `working_directory`, and only bound `target_source`/`primary_target` and `run_id`. Never search unlisted tools.
+If absent/failing: supplied schema-2 executable, child-only environment, immutable argv_prefix, declared arguments; `--skill ux-intent-definition`, language and working directory.
+For `--language`: Required presentation language for the latest natural-language user request as one well-formed BCP 47 tag. If the request explicitly asks for a response language, use that tag; otherwise use the dominant request language. Use en when mixed or ambiguous. A valid unsupported tag renders through the complete English pack. Missing or invalid input fails before governance evaluation.
+`target_source`: `explicit_target` if request names `primary_target`; `continued_target` if it unambiguously continues confirmed target; `current_repository` if request names this/current repo with one matching repo active. Otherwise omit the pair; cwd has no target authority.
+Quote shell values as data.
+For a result with `terminal: true`, the entire assistant response must consist only of host_action.text, copied verbatim. Add no question, explanation, heading, citation, link or other surrounding text; do not translate or reformat it; invoke no later tool and stop.
+On skill_continuation use only its target/control. Without that tool and a valid binding: `dispatcher_unavailable`; no runtime search, environment repair
+or help retries. Dispatch never authorizes.
+
 ## Authority
 
 This is an internal analytical skill, not a user gate. It cannot approve artefacts, grant
@@ -27,6 +59,7 @@ and approved as PRD content. Technical storage, derivation and component ownersh
 ## Inputs
 
 - approved UR and its intent, success signal, primary action and persona/context;
+- requested clarification deliverables, preview scenarios, interaction needs and timing in UR scope;
 - post-UR `delivery_context`, `ui_ux_impact`, reason and required flag;
 - existing behavior/ownership evidence for Brownfield work;
 - known working modes, state semantics, activation, blockers, recovery and transitions.
@@ -40,14 +73,73 @@ and approved as PRD content. Technical storage, derivation and component ownersh
    `primary_state_presentation_owner_by_mode`; do not turn either into technical ownership.
 5. Define activation/deactivation, blockers with visible next actions, actionable recovery (including
    visible retry for recoverable transient failure) and relevant state transitions.
-6. Propose observable PRD acceptance criteria and identify open product questions.
+6. Use UI previews as described below when seeing or trying the proposed behavior materially helps
+   resolve a product question. Propose observable PRD acceptance criteria and identify open questions.
 7. Return exactly one decision: `ready | blocked | not_applicable`.
 
+### UI previews for clarification
+
+Use a focused preview during UX intent clarification and reuse its findings during PRD alignment.
+Start with the user's task and unresolved decision, not a decorative dashboard. Show the relevant
+starting state, primary action, visible outcome and, where relevant, a blocker or recovery path.
+For example: can the user find what was checked, identify an open risk and decide whether to approve?
+Match known product context; identify illustrative data and unconfirmed assumptions explicitly.
+
+First preserve any preview explicitly required by approved UR scope, including its interaction,
+format and timing. Use an available visualization capability such as `visualize` when suitable.
+For an optional aid, a wireframe or written scenario walkthrough may answer the same question.
+For an explicitly requested interactive preview, unavailable tooling requires concrete recovery or
+an explicit scope revision; do not silently downgrade it or report the obligation fulfilled.
+Do not invent a preview requirement for already clear semantics, add a gate or install a dependency.
+Keep prototype interactions local; they cannot submit real approvals, modify delivery state or start
+implementation. A requested change to the actual product still follows its existing delivery path.
+
+#### Retain a run-bound preview
+
+For a preview used in this bound analysis, save a durable, reopenable copy under
+`.agdf/control/artefacts/<run_id>/previews/<short-name>-v<N>/index.html`, relative to the supplied
+governance target. Derive the run from the valid assignment, not cwd, recency or another run.
+Resolve containment and reject symlink escapes before writing; never put preview content in
+`runs/<run_id>/RUN_STATE.md`. Use a new unused version for revisions and preserve reviewed versions.
+Store a standalone document, preferably self-contained; bundle any local assets with relative
+references. For an inline visualization, keep its editable source in the host's permitted location
+and export a standalone copy to the run directory using that capability's supported flow. Remove
+host-only dependencies from the durable copy. Reopen it from its saved location and exercise the
+relevant interaction before claiming portability; an export or screenshot alone is insufficient.
+For required interactive PRD review, make that durable copy offline with local classic scripts,
+styles and images, without modules, remote resources or host calls. The PRD author owns the
+`Preview Review` descriptor and final criterion mapping; UX supplies the exact files and findings.
+
+Record the target run, source UR revision/digest, preview version, relative entry path, SHA-256 of
+the saved file (and local assets), covered scenarios and actual validation limits in the existing
+analysis `evidence` field. Keep observations tied to that version; a changed preview needs fresh
+inspection and affected user feedback. The UX report and PRD link the same evidence, not divergent
+copies. Use existing canonical recording/revision owners for their updates; preview bytes are
+supporting evidence, not a new approval authority or an automatically sealed artefact type.
+An unbound conversation preview remains advisory; do not create or select a run just to attach it.
+
+For PRD alignment, hand off a concise user review task, the readable versioned preview reference,
+covered scenarios and proposed criterion IDs alongside the observations. The existing presentation
+owner uses this material under interaction.md PRD Preview Review At Approval. Distinguish the
+proposed experience from delivered functionality and record any unavailable review access explicitly.
+
+Walk through the concrete task with the user and record observed behavior, feedback, unresolved
+assumptions and resulting proposed criteria in the existing `evidence`, `open_product_questions`
+and `proposed_prd_acceptance_criteria` fields. Distinguish agent inspection from user observation;
+check rendering and the relevant interaction before claiming they work. If no inspection or user
+observation occurred, say so. Retain the scenario and findings as well as the durable preview
+reference; chat-only or missing preview evidence cannot fulfill an explicit preview requirement.
+Preview feedback is analytical
+input, not PRD approval, implementation evidence or host qualification. Layout and component choices
+remain illustrative; only explicitly adopted product behavior becomes approved PRD content.
+
 ## Decision Rules
 
-- `ready`: every required output is reliable enough for PRD drafting.
+- `ready`: every required output is reliable enough for PRD drafting, including a requested preview
+  at the agreed fidelity, its durable evidence and resolution of material product questions.
 - `blocked`: user intent, modes, effective state, authority, activation, blockers, recovery, transitions
-  or Brownfield evidence is missing, contradictory or materially ambiguous.
+  or Brownfield evidence is missing, contradictory or materially ambiguous; or an explicitly required
+  preview, requested interaction, durable copy or decision-relevant observation remains missing.
 - `not_applicable`: routing proves that no relevant user-facing intent, state or recovery behavior changes.
 
 A conflict with approved UR routes to UR revision. A material product change discovered after PRD
@@ -83,7 +175,7 @@ Populate each field or use an explicit, justified `not_applicable`:
 
 For `blocked`, `blocking_reason`, `open_product_questions`, `affected_outputs`, `evidence`,
 `missing_evidence` and `required_next_step` are mandatory. When durable control state exists, write
-the result to `.agdf/control/artefacts/<key>/UX_INTENT_DEFINITION.md` without a Gate or approval field.
+the result to `.agdf/control/artefacts/<run_id>/UX_INTENT_DEFINITION.md` without a Gate or approval field.
 
 ## Forbidden
 
@@ -92,3 +184,5 @@ the result to `.agdf/control/artefacts/<key>/UX_INTENT_DEFINITION.md` without a
 - add a gate, approval value or implementation permission;
 - prescribe components, persistence, endpoints, styles or other technical design;
 - treat supporting analysis as authoritative beside the approved PRD.
+
+Apply interaction.md Bounded internal recovery and prepared observation for eligible bound follow-up. It adds no gate or approval authority.

package/generated/plugins/agdf/skills/ux-intent-definition/help.md

@@ -11,6 +11,9 @@
 Produces structured `ready | blocked | not_applicable` PRD input for user intent, success, working
 modes, effective state, visible ownership, activation, blockers, recovery and transitions. It is an
 internal non-authorizing analysis and fails closed when product semantics or evidence are unresolved.
+Preserves explicit preview commitments from UR scope, stores versioned reopenable previews in the
+bound run's artefact directory and maps observed findings to proposed PRD criteria. Optional aids
+can use a scenario walkthrough; a required interactive preview cannot silently become text.
 
 ## SEE ALSO
 

package/generated/plugins/agdf/submission/openai/availability.md

@@ -1,6 +1,6 @@
 # Availability Decision Record
 
-- release: `0.14.4`
+- release: `0.20.0`
 - decision_state: `pending`
 - decision_owner: verified publisher
 - countries: not selected

package/generated/plugins/agdf/submission/openai/capability-matrix.json

@@ -1,6 +1,6 @@
 {
   "schemaVersion": 1,
-  "releaseVersion": "0.14.4",
+  "releaseVersion": "0.20.0",
   "evidenceStates": [
     "common",
     "codex-specific",

package/generated/plugins/agdf/submission/openai/release-notes.md

@@ -1,4 +1,4 @@
-# AGDF 0.14.4 — Initial OpenAI Public Plugin Candidate
+# AGDF 0.20.0 — Initial OpenAI Public Plugin Candidate
 
 This candidate packages the canonical AGDF workflow skills for public discovery in ChatGPT and
 Codex. It helps users assess delivery fit, preserve explicit human approvals, inspect existing-system

package/generated/plugins/agdf/submission/openai/reviewer-cases.json

@@ -1,6 +1,6 @@
 {
   "schemaVersion": 1,
-  "releaseVersion": "0.14.4",
+  "releaseVersion": "0.20.0",
   "cases": [
     {
       "id": "positive-proportional-fit",