project-tiny-context-harness 0.8.1 → 0.8.3

package/README.md

@@ -62,7 +62,7 @@ No-install preview:
 
 ## Why It Exists
 
-`project_context/**` preserves small durable facts across sessions. The default workflow reads graph-relevant Context, supplements that route with one bounded Context search before `Context Delta`, and uses the platform's internal plan. For explicit long work, `long-task-delivery-v2` adds one complete Contract authority, compiled Source/REQ/CTRL/OBL/AC coverage, a one-time user model choice after Authority Lock, scoped progress and a source-recompiled Live Final Gate.
+`project_context/**` preserves small durable facts across sessions. The default workflow reads graph-relevant Context, supplements that route with one bounded Context search before `Context Delta`, and uses the platform's internal plan. For explicit long work, `long-task-delivery-v2` adds one complete Contract authority, fail-closed Source ownership, Control/applicability closure, semantic Counterfactuals, a one-time user model choice after Authority Lock, scoped progress and a protected-input-recompiled Live Final Gate.
 
 Minimal Context preserves durable facts, the Workflow Contract governs ordinary work, and the Long-Task Workflow adds explicit machine completion authority.
 
@@ -144,7 +144,7 @@ npm ci
 npm run smoke:quickstart
 npm run preview:pack
 cd /path/to/your/test-repo
-npm install -D /path/to/project-tiny-context-harness/tmp/ty-context/source-preview/package/project-tiny-context-harness-0.8.1.tgz
+npm install -D /path/to/project-tiny-context-harness/tmp/ty-context/source-preview/package/project-tiny-context-harness-0.8.3.tgz
 npx --no-install ty-context init --adopt
 make validate-context
 ```
@@ -207,17 +207,17 @@ For material UI, **UI Authority Closure** reconciles each stable surface/control
 
 One shared conditional purpose of both development paths is that Agent implementation, acceptance and testing fully conform to every material UI/UX fact selected design resources explicitly express within their declared scope and conditions. It activates only for a selected implementation handoff and does not infer unexpressed behavior or prove that the user supplied every desired requirement. Open Design can produce implementation-rich HTML/CSS/JS, specifications, tokens and assets, but capability is not a per-run guarantee: for a selected Web/App implementation handoff, `/design-resource-authoring` must explicitly commission and completely retrieve one machine-readable canonical entry plus its exact dependency closure, freeze every file with a digest and expose stable typed locators. Before `ready`, it exercises every declared verification method on those immutable bytes and blocks unresolved conflicts among code, specs, tokens and assets. That is source QA, not production acceptance. PNG may be a visual baseline, never the sole implementation source.
 
-The provider-neutral handoff is a residual semantic and index layer, not a textual copy of CSS. It closes every applicable subject × selected target × declared condition × UI/UX dimension cell across surface/flow, visual/content, component/control, state/interaction, motion, adaptation/input, accessibility and assets; it records explicit exclusions/not-applicable/unresolved meaning, Source Items, methods and blockers. Preflight resolves each typed HTML/Markdown/JSON/CSS locator against the declared immutable resource, verifies source/dependency closure and rejects unresolved or media-incompatible evidence. Exploration remains schema-free.
+The provider-neutral handoff is a residual semantic and index layer, not a textual copy of CSS. Its default unit is every complete observable design fact exposed by the acquired resource and supported inspector/oracle capability—not merely one Product Control—including addressable images, text, icons, component parts, smaller visual primitives, geometry, layout, style/token, content, state and behavior facts. It atomically indexes those facts, gives every resource a complete material-with-facts or honestly supporting-only inspection closure, and makes every applicable subject × selected target × declared condition × UI/UX dimension cell conserve the exact fact/evidence/Source/method sets across surface/flow, visual/content, component/control, state/interaction, motion, adaptation/input, accessibility and assets. An `exact_target` additionally requires full-target layout and pixel facts for every condition; otherwise the input stays a partial constraint or blocks. Preflight resolves typed locators against immutable resources, verifies source/dependency/fact closure and rejects unresolved, unsupported or media-incompatible evidence. Exploration remains schema-free.
 
-Those inputs remain ordinary Source. The default Workflow keeps exact task-local accounting of covered Source Items, declared verification methods, blockers, targets and conditions; every item must reach a production owner, cold-start journey and an executed final-candidate check whose failure remains attributable. Any unresolved, unmapped, unexecuted, stale or indistinguishable item blocks a complete claim and is reported as a gap. Long-Task projects the same sets into existing Claims, positive method Assertions, `surface_bindings`, typed evidence and Final Gate. These proof carriers are mutually exclusive: an active Long-Task never also runs the default closure. Generation success, screenshots, hashes and handoff preflight prove input completeness or integrity only.
+Those inputs remain ordinary Source. The default Workflow keeps exact task-local accounting of fact/Source/method/blocker/target/condition sets; every item must reach a production owner, cold-start journey and an executed final-candidate check whose failure remains attributable. Any unread, unsupported, unresolved, unmapped, unimplemented, unexecuted, stale or indistinguishable applicable fact blocks a complete claim and is reported as a gap. Long-Task projects the same sets into existing Claims, positive method Assertions, `surface_bindings`, typed fact-bound evidence and Final Gate. These proof carriers are mutually exclusive: an active Long-Task never also runs the default closure. Generation success, screenshots, hashes and handoff preflight prove input completeness or integrity only.
 
 The default Workflow performs UI Authority Closure and a conditional Design Authority Check before a material product, design, implementation or acceptance decision. It traverses affected stable keys to exactly one canonical adoption record, then actively opens every selected `exact-target` or `constraint`; a reference-index or handoff-index mention alone is not consumption. `DESIGN.md` canonically records project/system/component-family targets, while the owning Screen Contract records one-screen/interaction-specific targets. The canonical record owns interpretation, selection basis, readable immutable locator/digest, declared condition coverage and editable upstream owner/locator/update route; other layers keep only the stable key, canonical owner/anchor and local applicability. Missing, unreadable, stale or conflicting resources fail closed. Updates create a new immutable version instead of overwriting the adopted baseline. An unconfigured starter, candidate, style-only prose or inspiration does not authorize invented production layout, and a configured project visual system does not claim every page is implementation-ready. Explicit project design-system adoption routes to `/design-system-authoring`; standalone resource generation routes to `/design-resource-authoring`. Ordinary implementation with sufficient authority, local style fixes and throwaway prototypes remain lightweight.
 
-For a selected implementation handoff, both development paths first run `ty-context design-resource preflight <handoff.md>`. Incomplete acquisition, missing or undeclared dependencies, unsafe paths, stale digests, fictional locators, uncovered applicable cells, unsupported evidence and unresolved meaning all fail closed. Each workflow must still open the resources and prove the production implementation on the real entry.
+For a selected implementation handoff, both development paths first run `ty-context design-resource preflight <handoff.md>`. Incomplete acquisition, missing or undeclared dependencies, unsafe paths, stale digests, fictional locators, missing/unreferenced facts, false resource-fact closure, uncovered applicable cells, absent exact-target layout/pixel facts, unsupported evidence and unresolved meaning all fail closed. Each workflow must still open the resources and prove the production implementation on the real entry.
 
-For material work, `context_uiux_design` applies the projection above and keeps any risk-proportional coverage reasoning task-local. `context_development_engineer` traces every selected target/condition and the exact handoff sets through stable surface/control keys to the production route/component owner, cold-start real-user journey and independently attributable rendered/interactive checks. It inspects the first runnable production slice through the real entry before broad rollout and reports only combinations actually checked. Resource hashes, manifests and counts prove integrity only; an implementation screenshot cannot become its own target or implementation-conformance proof.
+For material work, `context_uiux_design` applies the projection above and keeps any risk-proportional coverage reasoning task-local. `context_development_engineer` traces every selected target/condition and the exact handoff sets through stable surface/control keys to the production route/component owner, cold-start real-user journey and independently attributable rendered/interactive checks. A first useful runnable production slice is a recommended real-entry feedback point when early localization is worth the cost, never an implementation gate; the final candidate always reruns the affected cold-start journey. Every declared/applicable combination remains covered—risk-only or pairwise sampling cannot replace it without authoritative scope narrowing or project-owned equivalence proof. Resource hashes, manifests and counts prove integrity only; an implementation screenshot cannot become its own target or implementation-conformance proof.
 
-An explicit Long-Task is the strong machine carrier of the same shared obligation. It resolves missing/conflicting UI authority before Compile, then preserves every applicable Control field through the existing projection: surface, region/location, type/label, user task, visibility/availability, trigger/input/validation/default, interaction/navigation, loading/empty/success/failure/recovery/permission/feedback and accessibility. Each non-empty field is an independent Source-backed Control Claim and protected product semantic; omitted fields create no Claim. Aggregated Product `surface_bindings` connect every Control to an owner surface, required product target, existing Technical route/component Bindings and a root-entry success journey. Selected exact/constraint targets bind frozen inputs and declared conditions to current actual/comparison artifacts through typed `design_conformance`; `verification_method_bindings` make every handoff method independently falsifiable, and every declared blocker preserves exact Source-item/method lineage into target-local machine proof or a target-blocking external confirmation. A blocker cannot be dismissed in-band: scope removal requires revised Source/Contract authority. Existing Claim, Assertion, Check, Stage, Binding, revision and Final Gate mechanisms are the sole Long-Task lifecycle and closure.
+An explicit Long-Task is the strong machine carrier of the same shared obligation. It resolves missing/conflicting UI authority before Compile, then closes all 22 canonical fields of every real Control through `field_coverage`: concrete `specified`, justified `not_applicable` or blocking `unresolved`; omission never means non-applicable. Specified and not-applicable entries create protected Control Claims at every declared applicability profile. `control_relation_closure` and `control_relations` also close shared-state, dependency/order, navigation, permission, validation, feedback and recovery chains across Controls and systems; not-applicable closure remains an exact negative Claim. Aggregated Product `surface_bindings` connect every Control to an owner surface, required product target, existing Technical route/component Bindings and a root-entry success journey. Selected exact/constraint targets bind frozen inputs and declared conditions to current actual/comparison artifacts through typed `design_conformance`; every `verification_method_binding` is independently falsifiable through its own typed `design_method` record and, per exact condition, exact handoff `fact_refs`, a declared method record and a method-native primary observation whose current path and digest cannot be reused by another cell. The compiled cell union equals the complete target fact set, and runtime evidence repeats the exact refs. Playwright requires explicit record/observation attachments from every project instance. Every declared blocker preserves exact Source-item/method/required-capability lineage into a machine Claim proved on the capability-qualified target or a target-blocking External Confirmation. A blocker cannot be dismissed in-band: scope removal requires revised Source/Contract authority. Existing Claim, Assertion, Check, Stage, Binding, revision and Final Gate mechanisms are the sole Long-Task lifecycle and closure.
 
 Combined design-and-implementation work may author candidates in ordinary Outcomes/Stages, but a candidate or planned target cannot authorize fidelity implementation. Selection must become real marked Context-reachable Source plus the owning Context/`DESIGN.md` reference and, after Authority Lock, an adopted Authority Revision. Browser visual ACs use `ui_browser`; a browser proxy, detached route or deep link cannot prove a native/root journey that can fail independently. Resource integrity and `visual_render` cannot satisfy selected-target implementation conformance. Frozen baselines are verifier inputs, generated actual renders/diffs are current artifacts, and subjective approval remains external. No `uiux_delivery` block, visual Claim type, resource registry, risk level, lifecycle state, Gate, required design directory, per-Control screenshot matrix or universal pixel threshold is added.
 
@@ -249,21 +249,27 @@ Maintainers may set `TY_CONTEXT_OPEN_DESIGN_MCP_COMMAND` plus optional `TY_CONTE
 
 ### Retired Source Plan Compatibility
 
-`/source-plan-authoring` is retained only as a compatibility pointer. `/long-task-workflow` opens the non-authoritative Contract Draft immediately and converges mixed-input inventory/synthesis, stable-key/control-level meaning, preference/research/delegation traceability, Source markers/provenance, acceptance/risk completeness and Contract mapping in one loop. Existing Source Plans remain ordinary Source; no standalone or internal Source-authoring stage, handoff, schema, gate, state or second plan is created.
+`/source-plan-authoring` is retained only as a compatibility pointer. `/long-task-workflow` opens the non-authoritative Contract Draft immediately and converges mixed-input inventory/synthesis, stable-key/Product Control-level meaning, preference/research/delegation traceability, Source markers/provenance, acceptance/risk completeness and Contract mapping in one loop. This semantic Control projection does not cap the separate complete-observable-design-fact inventory for selected resources. Existing Source Plans remain ordinary Source; no standalone or internal Source-authoring stage, handoff, schema, gate, state or second plan is created.
 
 ## Single-Goal Rolling Delivery
 
-The explicit Long-Task Workflow uses one platform-native Goal, one user-selected repository/workspace, one complete `long-task-delivery-v2` Contract and one Final Gate. Outcomes are independently decidable acceptance units; Delivery Set orchestration and top-level Contract splitting inside one selected delivery are retired.
+The explicit Long-Task Workflow uses one platform-native Goal, one user-selected repository and verification/convergence workspace, one complete `long-task-delivery-v2` Contract and one Final Gate. Outcomes are independently decidable acceptance units; Delivery Set orchestration and top-level Contract splitting inside one selected delivery are retired.
 
-Raw/revised proposals, selected design resources and mixed attachments enter one Source-bound Contract Draft loop immediately. Complete input coverage, stable keys, control-level meaning, acceptance/risk, direct/derived/delegated/evidence-backed provenance, Source markers and Contract mapping converge together. Unknown decision-changing preferences still trigger one targeted clarification before Preflight/Compile can succeed; defensible recommendations are written into real Source rather than hidden in YAML, while high-risk actions remain external confirmations. Legacy Source Plan structure never blocks authoring.
+Its proof claim is conditional: complete and accurate Source at the declared observable granularity, meaning-preserving projection, complete applicability expansion and a semantically sound named project oracle/trusted verifier boundary make `AcceptedDeliveryTerminal`—exactly a fresh `machine_accepted` result with no pending External Confirmation—imply that no declared observable drift remains. `machine_accepted_external_pending` proves only the machine-verifiable declared scope; full delivery remains qualified and Harness does not complete the native Goal. Harness enforces and freezes many premises, but it cannot discover undeclared requirements or prove an arbitrary oracle truthful.
+
+Raw/revised proposals, selected design resources and mixed attachments enter one Source-bound Contract Draft loop immediately. Complete input coverage, stable keys, Product Control-level meaning, selected-resource design facts, acceptance/risk, direct/derived/delegated/evidence-backed provenance, Source ownership and Contract mapping converge together. Every non-empty line in declared Markdown Source belongs to one Material `ty-source-item`, the single validated `design-resource-handoff-v1` formal block or a closed-grammar background block containing only text-free anchors/horizontal rules or fixed-field `ty-source-provenance` comments. Text-bearing headings, free-form provenance, arbitrary background prose and unclassified text fail closed unless classified as Material. At least one marked technical obligation carries `aspect=architecture` and maps to an independently provable architecture obligation. Unknown decision-changing preferences still trigger one targeted clarification before Preflight/Compile can succeed; defensible recommendations are written into real Source rather than hidden in YAML, while high-risk actions remain external confirmations. Legacy Source Plan structure never blocks authoring.
 
 Before the first successful formal Compile, `delivery-contract.yaml` is one non-authoritative Contract Draft. `/long-task-workflow` opens it at entry and revises the same Draft across Source refinement, repository/Context reads, mapping and Preflight repairs; a complete Contract need not fit one response. Source completeness is a convergence condition for Preflight/Compile, not a prior phase. There is no standalone Contract Draft Skill or Authoring State.
 
 The Long-Task Skill keeps objective/boundary/activity routing in its main file and loads one-level Source-bound Draft/Contract-mapping, evidence-design and authority-lifecycle references as applicable. Draft input repair and Contract mapping are concurrent activities, not serial phases. This is instruction packaging only, not a second authority. It performs the shared Architecture Deliberation during Draft authoring. Declared architecture invariants use existing obligations/constraints/forbidden shortcuts, owner/path/Binding boundaries and project-owned executable Checks; a functional AC cannot substitute for an independently failing architecture claim. Final Gate is the sole Long-Task Architecture Conformance carrier.
 
-A Draft Outcome is simply an Outcome before Authority Lock. Outcomes decompose independently observable, decidable and target-verifiable results to improve dependency-ready implementation, targeted verification, failure localization, resume and stale-result invalidation. `depends_on` means acceptance readiness and the Rolling Frontier is temporary; an Outcome is not a Worker, scheduler task, queue or parallel unit. Outcome decomposes execution and diagnosis, not completion authority, so one complete current-snapshot Final Gate remains mandatory.
+A Draft Outcome is simply an Outcome before Authority Lock. Outcomes decompose independently observable, decidable and target-verifiable results to project acceptance/verification readiness, localize failures, resume findings and invalidate stale results. `depends_on` means acceptance and intermediate-proof readiness, not implementation permission, and the Rolling Frontier is a temporary advisory projection. The current Goal chooses implementation order and may optionally use one or multiple platform-native agents/subagents; Harness creates no delegation runtime, scheduler, state or proof, agent reports are non-authoritative, and all accepted output converges into the selected verification workspace. An Outcome is not a Worker, scheduler task, queue or parallel unit. Outcome decomposes diagnosis and proof ownership, not completion authority, so one complete current-snapshot Final Gate remains mandatory.
+
+When a declared result can pass on a proxy surface while failing in its target runtime, the earliest owning Outcome declares a project-owned Check that exercises the target during the current Check execution. A tracked report, screenshot, binary, log or historical run cannot be the sole runtime proof. Global/Outcome applicability profiles bind exact target, journey, atomic duplicate-free dimensions, Given condition/input/state refs and ordered When refs; every Claim-bearing Assertion proves one Claim in one matching profile, and every actual applicability × proof-surface cell remains attributable without risk-based, pairwise or sampled substitution. Required product targets declare their family, cold-start and production-root capabilities. Every behavioral Claim-bearing Assertion also has a same-Check claim-local `replace_json_value` or `replace_text` Counterfactual that keeps a claimless target-runtime liveness Assertion passing while wrong behavior fails; whole-file replacement cannot prove semantic binding. Population binds a real universe carrier into its owning Check snapshot and proves universe = eligible = observed plus valid exclusions. Runner identity recursively freezes the supported direct-literal local verifier module/config/data graph and rejects non-literal loaders/`createRequire`; package scripts need a recoverable static Node entry. Candidate/runtime material remains explicitly owned by `input_paths`, `expected_output_paths` or `artifact_globs`; other indirect Oracle access belongs in `verification_inputs` or the named trusted boundary. Author each Check's `input_paths`/Bindings as its smallest sound invalidation envelope and keep every Counterfactual carrier traceable from the declared target root. The first useful runnable boundary and later coalesced changes are recommended targeted-feedback points only when early localization is worth the cost. Refresh `progress_stale` only before an intermediate decision relies on that result. Implementation and Final Gate may proceed with stale or absent Progress because the Gate ignores it and reruns every Check. `verify --explain` previews bounded declared runner invocations without execution or Progress writes, but cannot predict duration or runner-internal subprocesses. This adds no generic reachability claim, implementation gate, second executing diagnose mode, scheduler, trigger queue, `platform_impact` flags, completion state or per-edit rebuild rule; runtime-specific readiness/build/process behavior stays in the project runner, and Final Gate remains authoritative.
+
+Long-Task Anti-Degradation Assurance protects current causal-chain truth, cross-version interception strength and the adjacent `F = Implementation Freedom Boundary`. Current-implementation Context must match the indexed code/runtime, which must still realize the two necessary meaning-capture and fail-closed final-proof responsibilities under the explicit theorem boundary. `F` is an efficiency/anti-process-bloat invariant, not a third responsibility or theorem premise: within Source/Contract, architecture, safety, forbidden-shortcut and irreversible/external-action boundaries, implementation order, methods, feedback cadence and optional one-agent or multi-agent/subagent execution remain Goal-owned. Harness adds no development phase/method Gate, per-edit mandate, agent scheduler/state or delegation proof. Weakening the protected purpose, key logic, either responsibility, theorem boundary or `F` requires an explicit project-owner design-purpose decision and replacement proof; a new development-stage constraint must also close a distinct path that final proof or a lighter project-owned check cannot cover and have positive net ROI. Existing Context, indexes, tests, critical sentinels, routing and parity gates carry this assurance without another Authority, Gate or state; they cannot infer omitted requirements or guarantee immutability against deliberate fully authorized joint weakening.
 
-When a declared result can pass on a proxy surface while failing in its target runtime, the earliest owning Outcome declares a project-owned Check that exercises the target during the current Check execution. A tracked report, screenshot, binary, log or historical run cannot be the sole runtime proof. Author each Check's `input_paths`/Bindings as its smallest sound invalidation envelope and keep every Counterfactual carrier traceable from the declared target root. Run that Check at the first useful runnable boundary; later `progress_stale` only reports that prior evidence no longer covers current inputs. Coalesce related edits, use the cheapest reliable feedback, and refresh before dependent reliance or Final Gate. `verify --explain` previews bounded declared runner invocations without execution or Progress writes, but cannot predict duration or runner-internal subprocesses. This adds no generic reachability claim, second executing diagnose mode, scheduler, trigger queue, `platform_impact` flags, completion state or per-edit rebuild rule; runtime-specific readiness/build/process behavior stays in the project runner, and Final Gate remains authoritative.
+Workflow changes then pass two hard gates in order: first prove `Coverage_new ⊇ Coverage_old`, `FalseNegative_new ⊆ FalseNegative_old` and preserve non-bypassable Authority, fail-closed behavior and complete-current-final-snapshot proof; if that cannot be proved, preserve the current formal acceptance path. Only then require incremental anti-degradation or purpose-fulfillment benefit to exceed all authoring, runtime, state, recovery, maintenance, test, process, introduction and migration cost. Cost never compensates for weaker drift detection. Positive net ROI means the proposal may be considered, not that it is automatically adopted.
 
 ### One-time execution-model choice
 
@@ -306,19 +312,19 @@ ty-context long-task close <workdir>
 ty-context long-task abandon <workdir> [--force-corrupt-state]
 ```
 
-Compact authoring omits only deterministic defaults and normalizes identically to the expanded form. `preflight` is a read-only aggregated Source/REQ/CTRL/OBL/AC and repository check. Before first Authority Lock, Preflight and direct Compile both classify every HEAD-relative changed path as protected, expected change, allowed support, forbidden or unclassified; forbidden and unclassified paths block, so Compile cannot bypass Preflight. During first enable, only exact current package-asset files for configured managed destinations plus exact config/hook files are temporarily protected; managed directory roots and broad `.codex/**` are never exempt. Compile then generates Global plus Outcome Result/Requirement/Control-field/Non-completing/Technical Claims, rejects uncovered Claims and makes the first successful formal Compile the Authority Lock. Every Compile result includes a lifecycle event, `delivery_completed_by_this_event: false`, `native_goal_effect: none` and a next action. The first Compile result emits `execution_model_checkpoint.required: true` plus its terminal-turn/explicit-choice contract; later Compile revisions emit `required: false`. Every later authority change still compares with active authority regardless of progress, Receipt/cache deletion or implementation restoration. Source/Context/Product/Acceptance/Global/verifier content, resolved runners and verification inputs are frozen in the common-dir Active Authority V3 record.
+Compact authoring omits only deterministic defaults and normalizes identically to the expanded form; full Context remains explicit and mandatory. `preflight` is a read-only aggregated closed-grammar Source/background and architecture obligation, REQ/CTRL field-and-relation closure, OBL/AC, atomic applicability, Population universe, target and blocker capability, per-method design record/primary-observation artifact, supported direct-literal verifier dependency, narrow semantic witness and repository check. Before first Authority Lock, Preflight and direct Compile both classify every HEAD-relative changed path as protected, expected change, allowed support, forbidden or unclassified; forbidden and unclassified paths block, so Compile cannot bypass Preflight. During first enable, only exact current package-asset files for configured managed destinations plus exact config/hook files are temporarily protected; managed directory roots and broad `.codex/**` are never exempt. Compile then generates Global plus Outcome Result/Requirement/Control-field/Control-relation/Non-completing/Technical Claims at exact applicability, rejects uncovered cells and makes the first successful formal Compile the Authority Lock. Every Compile result includes a lifecycle event, `delivery_completed_by_this_event: false`, `native_goal_effect: none` and a next action. The first Compile result emits `execution_model_checkpoint.required: true` plus its terminal-turn/explicit-choice contract; later Compile revisions emit `required: false`. Every later authority change still compares with active authority regardless of progress, Receipt/cache deletion or implementation restoration. Source/full Context/Product/Acceptance/Global/verifier content, resolved runners and recursively frozen verification inputs are frozen in the common-dir Active Authority V3 record.
 
 `diagnose-revision` performs a side-effect-free candidate Compile and only exercises existing active Check identities whose runner/verifier authority is unchanged. Its output explicitly denies acceptance, Progress and pending-state writes. `compile --revise` auto-adopts mechanically bounded revisions; for a user-decision revision it emits `authority_revision_pending`, the exact decision id, deterministic material summary, `user_decision_reasons` and the self-contained human `decision_brief` before failing closed. Approving a different or stale id is rejected. Adoption emits `authority_revision_adopted` and returns to rolling execution rather than completion.
 
-Targeted verify rechecks active task/revision/compiled/worktree identity and applies the same workspace categories against the immutable baseline before writing scoped Progress. `verify --explain` groups selected Main Raw Executions, lists applicable Counterfactual calls and declared retry bounds, but runs nothing and writes no Progress. Counterfactual Findings first enter the owning Check Result, invalidate an otherwise passed Check, clear Claim Proofs and remain visible in status/resume; Global Checks reuse the same Progress type without a Global Outcome state. Final Gate repeats the identity check after all Checks; Stop/close clear only the accepted identity through CAS. Commit, migration, clear and abandon share one active-state lock. `abandon --force-corrupt-state` is reserved for corrupt continuity or stale lock cleanup and preserves Contract, Source, Context and Git content.
+Targeted verify rechecks active task/revision/compiled/worktree identity and applies the same workspace categories against the immutable baseline before writing scoped Progress. `verify --explain` groups selected Main Raw Executions, lists applicable Counterfactual calls and declared retry bounds, but runs nothing and writes no Progress. Counterfactual Findings first enter the owning Check Result, invalidate an otherwise passed Check, clear Claim Proofs and remain visible in status/resume; Global Checks reuse the same Progress type without a Global Outcome state. Final Gate first rejects stale accepted inputs, then recompiles and hashes Contract/fragments, Source, full Controlling Context, verifier/runner, recursively frozen local dependencies, verification inputs and workdir inputs before and after all Checks; any protected-input or Active Authority race fails closed. Stop/close clear only the accepted identity through CAS. Commit, migration, clear and abandon share one active-state lock. `abandon --force-corrupt-state` is reserved for corrupt continuity or stale lock cleanup and preserves Contract, Source, Context and Git content.
 
-`status` and read-only `resume` report the current fresh Final Receipt as `final_workflow_status` (or `null` after drift) plus the active Contract's complete `external_confirmations`. `progress_passing` is targeted repair evidence rather than “Outcome complete”; `progress_stale` is a freshness fact rather than a current pass or immediate rerun command, and `final_workflow_status: null` means unfinished. Every accepted Stop emits one non-blocking terminal-scope `systemMessage`; external-pending results also name every confirmation. Final/Stop/close report `acceptance_scope: declared_machine_authority` and `native_goal_effect: none`; close also reports `closed_scope: machine_authority`. Before platform-native Goal completion, the Agent performs a veto-only Goal/user-to-Source conformance review that cannot create proof. `status: closed` means only that machine Authority was cleared, not that the native Goal or external delivery completed.
+`status` and read-only `resume` report the current fresh Final Receipt as `final_workflow_status` (or `null` after drift) plus the active Contract's complete `external_confirmations`. Their Frontier/next-action output is advisory verification/repair guidance; legacy `ready_for_implementation` is only a compatibility alias and never gates implementation order. `progress_passing` is targeted repair evidence rather than “Outcome complete”; `progress_stale` is a freshness fact rather than a current pass or immediate rerun command, and `final_workflow_status: null` means unfinished. Every accepted Stop emits one non-blocking terminal-scope `systemMessage`; external-pending results also name every confirmation. Final/Stop/close report `acceptance_scope: declared_machine_authority` and `native_goal_effect: none`; close also reports `closed_scope: machine_authority`. Before platform-native Goal completion, the Agent performs a veto-only Goal/user-to-Source conformance review that cannot create proof. `status: closed` means only that machine Authority was cleared, not that the native Goal or external delivery completed.
 
-New authoring uses inline Outcomes. Existing `outcome_files` remains physical compatibility only and creates no semantic or completion boundary. A Long Task requires real Source, and every declared Source file contains at least one Material Item; background-only references remain outside Source Authority. Every Material Source Item is wrapped in the original Markdown with a non-rendering, uniquely keyed `ty-source-item:start/end` marker; `control` is a first-class kind, marker keys and `source_claim` keys are set-equal, and statements are text-exact after limited whitespace normalization. Every non-decision Source item owns one same-kind, same-text canonical target and duplicate ownership fails. Outcome Source Acceptance maps to criterion-identical `<outcome>.<check>.<assertion>` with an independently Source-backed non-Result Claim; Global Source Acceptance maps to criterion-identical `GLOBAL.<check>.<assertion>`, proves no Outcome Claim and needs an independently Source-backed Global Claim. Typed dispositions keep Requirements, Controls, Acceptance, Results, Fact/Affected-Outcome Risk, Non-goals, External Confirmations and Decisions distinct; `out_of_scope` is retired. Ordinary prose remains valid after marker-only enumeration.
+New authoring uses inline Outcomes. Existing `outcome_files` remains physical compatibility only and creates no semantic or completion boundary. A Long Task requires real Source, and every declared Source file contains at least one Material Item. Every Material Source Item is wrapped in the original Markdown with a non-rendering, uniquely keyed `ty-source-item:start/end` marker; at least one technical obligation has `aspect=architecture`. Other non-empty lines may occur only in the validated handoff or a closed-grammar Markdown-structure/provenance background block. Arbitrary background prose and unclassified, nested, overlapping, unclosed or empty sections fail closed. `control` is a first-class kind, marker keys and `source_claim` keys are set-equal, and statements are text-exact after limited whitespace normalization. Every non-decision Source item owns one same-kind, same-text canonical target and duplicate ownership fails. Outcome Source Acceptance maps to criterion-identical `<outcome>.<check>.<assertion>` with an independently Source-backed non-Result Claim; Global Source Acceptance maps to criterion-identical `GLOBAL.<check>.<assertion>`, proves no Outcome Claim and needs an independently Source-backed Global Claim. Typed dispositions keep Requirements, Controls, Acceptance, Results, Fact/Affected-Outcome Risk, Non-goals, External Confirmations and Decisions distinct; `out_of_scope` is retired. Complete syntactic ownership cannot prove that the user expressed every real requirement or that marked Source is factually accurate; those remain explicit upstream premises.
 
 After Authority Lock, semantic/Product Claim/Acceptance/verifier-kernel changes and proof weakening require an exact user decision. Mechanically bounded implementation repairs and raw snapshot changes that preserve locked meaning auto-revise but still invalidate affected evidence. Pure package root/version relocation auto-revises; schema/hook byte changes do not. Contract and Check execution field policies prevent new fields from bypassing authority or raw-execution identity. Every path-bearing field uses one canonical grammar: Windows separators and one leading `./` normalize, while internal `.`/`..`, controls, absolute/drive/UNC paths and unsupported glob syntax fail closed.
 
-Supported runners: `package_script`, `project_binary`, `node_oracle`, `playwright_test`.
+Supported runners: `package_script`, `project_binary`, `node_oracle`, `playwright_test`. Package scripts require a recoverable static Node entry for formal acceptance.
 
 Supported proof surfaces: `ui_browser`, `runtime_behavior`, `api_contract`, `data_state`, `security_boundary`, `population_coverage`, `implementation_structure`.
 
@@ -330,7 +336,7 @@ L0 local work stays on the default workflow. L1 standard long work uses the Deli
 
 Users may raise risk to strict. Explicit `standard` below the computed floor fails. Strict negative, counterfactual, population, security, environment and rollback/recovery proof is compiler-enforced as applicable. Scope escape returns a `scope_escape` Finding for revision and recompilation in the same Goal.
 
-Agent prose, a command exit code, handwritten state, historical targeted passes and missing/weak proof cannot create accepted. Evidence adapters derive from runner kind: only `playwright_json_v1` from `playwright_test` may prove `ui_browser`; other runners produce `structured_json_v2`. Every Outcome has a non-Result atomic Claim and all required surfaces must be non-empty, unique and covered. Across every Check sharing one Raw Execution identity, a Claim-bearing Observation is unique to one Assertion. Playwright Claim evidence is only `playwright.case.<ac>.passed equals true`; `[ac:<key>]` binds one declared AC per Test, ordinary tags are ignored, and missing/skipped/flaky/unexpected/timed-out/interrupted/multi-AC/duplicate-per-project evidence fails closed while distinct projects aggregate all-of. Structured Counterfactuals require exit zero; weak Playwright Counterfactuals may accept exit one only when every unexpected Test Instance is exactly a designated executed AC and no root/unbound/extra/timeout/interruption/flaky or other Evidence failure exists. Ordinary Playwright Baselines still require exit zero, and report/instance diagnostic observations cannot prove Claims. Each Claim-bearing structured Check needs same-Check, Claim-related Counterfactual sensitivity; unrelated Artifacts/Checks do not count, Population exempts only its same-Check Claims except under weak observability, and Result sensitivity needs a related non-Result root. Claim/Population proofs are emitted only for a fully passed Check. Findings and Explain trace Source, canonical target, Claim, AC/criterion, required surfaces, Check, adapter, Observation and owner paths.
+Agent prose, a command exit code, handwritten state, historical targeted passes and missing/weak proof cannot create accepted. Evidence adapters derive from runner kind: only `playwright_json_v1` from `playwright_test` may prove `ui_browser`; other runners produce `structured_json_v2`. Every Outcome has a non-Result atomic Claim and all required surfaces must be non-empty, unique and covered. Each Claim-bearing Assertion proves exactly one Claim at one matching atomic applicability profile. Across every Check sharing one Raw Execution identity, a Claim-bearing Observation is unique to one Assertion. Playwright Claim evidence is only `playwright.case.<ac>.passed equals true`; `[ac:<key>]` binds one declared AC per Test, ordinary tags are ignored, and missing/skipped/flaky/unexpected/timed-out/interrupted/multi-AC/duplicate-per-project evidence fails closed while distinct projects aggregate all-of. Every behavioral Claim-bearing Assertion requires same-Check claim-local `replace_json_value` or `replace_text` sensitivity plus a claimless target-runtime liveness Assertion that remains passing; `replace_file` is compatibility-only. Population proves a bound universe exactly but never waives this witness. Structured Counterfactuals require exit zero; weak Playwright Counterfactuals may accept exit one only when every unexpected Test Instance is exactly a designated executed AC and no root/unbound/extra/timeout/interruption/flaky or other Evidence failure exists. Ordinary Playwright Baselines still require exit zero, and report/instance diagnostic observations cannot prove Claims. Unrelated Artifacts/Checks do not count. Claim/Population proofs are emitted only for a fully passed Check. Findings and Explain trace Source, canonical target, Claim/applicability, Assertion, required surfaces, Check, adapter, Observation and owner paths.
 
 ## Upgrade And Compatibility
 
@@ -369,7 +375,7 @@ make validate-harness
 
 The modularity gate is `ty-context check-modularity`. Scoped waivers require `owner`, `introduced_at`, `reason`, `tracking_issue` and `expiry_condition`.
 
-The synchronized local preview tarball is named `project-tiny-context-harness-0.8.1.tgz`.
+The synchronized local preview tarball is named `project-tiny-context-harness-0.8.3.tgz`.
 
 ## Community And Further Reading
 
@@ -388,7 +394,7 @@ For concrete examples, see the [fresh-agent recovery walkthrough](https://github
 
 ## Honest Limits
 
-Tiny Context does not create or restore a platform Goal, prove that every requirement was declared, guarantee bounded keyword search finds every synonym or indirect dependency, switch the host-selected model, provide core parallel mutation, observe platform tokens/model calls, or own Git/PR/CI/deployment/human product confirmation. The installed package verifier and Git metadata are trusted; external platforms own network isolation, and deliberate same-user/admin tampering remains outside the local threat model.
+Tiny Context does not create or restore a platform Goal, prove that every requirement was declared, guarantee bounded keyword search finds every synonym or indirect dependency, switch the host-selected model, provide a parallel mutation/delegation runtime, observe platform tokens/model calls, or own Git/PR/CI/deployment/human product confirmation. A platform Goal may use opaque implementation delegation, but Harness does not persist or accept it as evidence. The installed package verifier and Git metadata are trusted; external platforms own network isolation, and deliberate same-user/admin tampering remains outside the local threat model.
 
 ## License