@codyswann/lisa 4.74.9 → 4.75.0

package/plugins/lisa-devin/skills/lisa-linear-validate-issue/SKILL.md

@@ -0,0 +1,504 @@
+---
+name: lisa-linear-validate-issue
+description: "Validates a proposed Linear work item spec (Project, Issue, or sub-Issue) — or an existing Linear item — against the organizational quality gates without writing anything. Returns a structured PASS/FAIL report per gate with concrete remediation. Single source of truth for what makes a valid Linear item — both the write path (linear-write-issue runs it pre-write) and the dry-run path (linear-to-tracker runs it during PRD intake) call this skill so the bar can never drift."
+allowed-tools: ["Skill", "Bash"]
+---
+
+# Validate Linear Work Item: $ARGUMENTS
+
+Run all organizational quality gates against a Linear work item spec OR an existing Linear item. **This skill is read-only — it never writes to Linear.** The output is a structured report consumed by callers (`lisa-linear-write-issue` for pre-write gating, `lisa-linear-to-tracker` for PRD dry-run, `lisa-linear-verify` for post-write checks).
+
+## Configuration
+
+Reads `linear.workspace`, `linear.teamKey` from `.lisa.config.json` (with `.local` override) for any feasibility lookups.
+
+## Input
+
+`$ARGUMENTS` is one of:
+
+1. **An existing Linear identifier** (e.g. `ENG-123` for an Issue, or `<workspace>/project/<slug>-<id>` for a Project): fetch and validate the live state.
+2. **A proposed item spec** (YAML block, see schema below): validate as-is without touching Linear.
+
+### Standalone entry point — validating an item written by another path
+
+Input form 1 is a **supported entry point in its own right**, not only an internal step of a
+caller flow. Point this skill at any existing item — however it was written, including by a
+bespoke script, a direct API or GraphQL call, or the vendor's own web UI — and it fetches the
+live state and runs the full gate set against it.
+
+Copy-pasteable, via the Skill tool:
+
+```text
+Skill(skill: "lisa-linear-validate-issue", args: "ENG-123")
+```
+
+where the argument is an Issue identifier such as `ENG-123`, or a Project URL. The report it returns is the same structured PASS/FAIL
+report the write path consumes, so a bespoke write path can discharge both the pre-write
+validate and the post-write verify obligation with it (see the bespoke-path section of
+`lisa-linear-write-issue`).
+
+This skill is plugin-resident. It is invoked through the Skill tool and is **not** expected to
+appear in any repository's `scripts/` directory; not finding a shell script by this name is
+not evidence that the capability is absent.
+
+### Spec schema
+
+```yaml
+issue_type: Story          # Epic | Story | Task | Bug | Spike | Sub-task | Improvement
+team_key: ENG
+summary: "[CU-1.2] Upload contract PDF from settings"
+priority: 3                # Linear native: 0=No priority, 1=Urgent, 2=High, 3=Medium, 4=Low
+parent_project_id: <uuid>  # Required for Story/Task/Improvement when in Epic context
+parent_issue_id: <uuid>    # Required for Sub-task (the Story Issue)
+description: |             # Full description text — every required section
+  ## Context / Business Value
+  ...
+
+  ## Technical Approach
+  ...
+
+  ## Acceptance Criteria
+  1. Given <precondition>
+     When <action>
+     Then <observable outcome>
+
+  ## Out of Scope
+  ...
+
+  ## Target Backend Environment
+  dev
+
+  ## Sign-in Required
+  Account: ...
+
+  ## Repository
+  backend-api
+
+  ## Validation Journey
+  ...
+
+# Behavioral flags — caller asserts these so the validator picks the right gates
+# (on a LIVE item `runtime_behavior_change` is DERIVED from the stored declaration, not taken on assertion — S8)
+runtime_behavior_change: true     # → requires Target Backend Environment + Validation Journey; LIVE items derive it from the stored declaration instead (S8)
+authenticated_surface: true       # → requires Sign-in Required
+artifacts_attached: true          # → requires Source Precedence section
+relations: [{ id: "ENG-99", type: "blocked_by" }]   # known issue relations (may be empty)
+remote_links: [{ url: "https://github.com/...", title: "PR #42" }]
+build_ready: true                 # caller asserts the build-ready role (the `Ready` state) is/would be set — see S15
+child_refs: ["ENG-601", "ENG-602"]   # known child work (sub-issues / project-member issues / blocked_by parentage) — see S15
+prd_source: "https://notion.so/..."    # set when the Issue was generated from a PRD — requires the Source Requirement section, see S16
+```
+
+If the caller passes only an identifier, fetch the item via `lisa-linear-access operation: get-issue` (Issue) or `lisa-linear-access operation: get-project` (Project), derive the same fields from the fetched data — including `runtime_behavior_change` (derived from the `## Target Backend Environment` declaration per `derived-branch-plan`, and authoritative over any caller assertion), `build_ready` (the Issue's state is the configured `ready` state) and `child_refs` (sub-issues, project-member issues, plus `blocked_by` parentage, resolved as in `lisa-linear-read-issue`) so S15 can classify the item — then run gates.
+
+For existing items written through another path, use the [standalone entry point](#standalone-entry-point--validating-an-item-written-by-another-path). A transport read-back does not prove the stored specification passes the gates.
+
+## Gates
+
+Gates are grouped into **Specification** (spec-only checks, no Linear lookups) and **Feasibility** (requires Linear lookups). The dry-run path may opt to run Specification gates only; the write path runs both.
+
+Each gate is tagged with a fixed `category` and a `product_relevant` boolean. Categories drive how downstream callers (notably `lisa-linear-prd-intake`) translate failures into product-facing comments; `product_relevant=false` failures indicate internal data-quality problems the agent should fix itself rather than ask product to clarify.
+
+Per-type content requirements are defined once in the vendor-neutral `work-item-definition-of-ready` rule (eager + reference); gates S4–S6 and S17 enforce them, and S18 enforces the stateless-pickup property directly.
+
+| Gate | Category | Product-relevant |
+|------|----------|------------------|
+| S1 Required core fields | `structural` | false |
+| S2 Summary format | `structural` | false |
+| S3 Description three audiences | `product-clarity` | true |
+| S4 Acceptance criteria in Gherkin | `acceptance-criteria` | true |
+| S5 Bug-specific content | `product-clarity` | true |
+| S6 Spike-specific content | `scope` | true |
+| S7 Project parent declared | `structural` | false |
+| S8 Target Backend Environment | `technical` | false |
+| S9 Sign-in Required | `technical` | false |
+| S10 Single-repo scope | `scope` | false |
+| S11 Validation Journey | `acceptance-criteria` | true |
+| S12 Source Precedence | `design-ux` | true |
+| S13 Relationship Search | `dependency` | true |
+| S14 Evidence manifest binding (leaf work units) | `acceptance-criteria` | true |
+| S15 Leaf-only build-ready | `structural` | false |
+| S16 Source Requirement traceability | `product-clarity` | true |
+| S17 Improvement measurability | `acceptance-criteria` | true |
+| S18 Stateless-pickup dry-run | `product-clarity` | true |
+| S19 Branch Plan derivation | `technical` | false |
+| S20 Named-control reachability | `acceptance-criteria` | true |
+| F1 Issue type valid in team | `structural` | false |
+| F2 Project parent exists and is in same team | `structural` | false |
+| F3 Linked items exist | `structural` | false |
+| F4 Required labels exist (or can be created) | `structural` | false |
+| F5 Required external access provable | `technical` | true |
+
+Category values are the same fixed set as `lisa-jira-validate-ticket`:
+
+- `product-clarity` — feature behavior or user intent unclear in the PRD
+- `acceptance-criteria` — pass/fail conditions missing or ambiguous
+- `design-ux` — visual or interaction spec missing
+- `scope` — boundary unclear, items overlap, split needed
+- `dependency` — blocked by another team / system / decision
+- `data` — data source / shape / volume unspecified
+- `technical` — engineering decision required
+- `structural` — internal data-quality problem the agent must fix itself, not surface to product
+
+### Specification Gates
+
+#### S1 — Required core fields
+
+`team_key`, `issue_type`, `summary`, `priority`, `description` must all be present and non-empty.
+
+#### S2 — Summary format
+
+- Single line, ≤ 100 characters
+- Imperative voice ("Add X", "Fix Y", not "Adding X" or "X is broken")
+- Bug / Task / Sub-task summaries SHOULD start with a `[repo-name]` prefix when the project convention uses one
+
+#### S3 — Description has all three audiences
+
+Description text must include all of these sections (case-insensitive `##` markdown headings):
+- `Context / Business Value` — stakeholder-facing
+- `Technical Approach` — developer-facing
+- `Acceptance Criteria` — coding-assistant-facing
+- `Out of Scope` — explicit non-coverage list
+
+Missing any → FAIL with name of missing section.
+
+#### S4 — Acceptance criteria in Gherkin
+
+Applies when `issue_type ∈ {Story, Task, Bug, Sub-task, Improvement}`.
+
+The `Acceptance Criteria` section must contain at least one criterion in `Given / When / Then` form. Reject prose-only criteria, "should work" language, or numbered lists without Given/When/Then verbs.
+
+#### S5 — Bug-specific content
+
+When `issue_type = Bug`, description must additionally include the full bug anatomy from the `work-item-definition-of-ready` rule:
+
+- **Agent-executable reproduction** — numbered steps a stateless agent can run mechanically: exact entry point, named account/role (consistent with Sign-in Required when authenticated), concrete data state, exact actions. Human-followable-only prose ("click around until it breaks") FAILs. A linked failing test satisfies this outright and is the preferred form.
+- **Expected vs. actual behavior**, naming or linking the *source* of "expected" (spec, PRD requirement, prior release behavior) — a fix target is a fact, not an opinion.
+- **Environment + version** — where reproduced and the build/commit observed; last-known-good when known (`unknown` must be stated, not omitted).
+- **Reproducibility rate** — always, or intermittent with observed frequency; intermittent invalidates run-once verification.
+- **Occurrence evidence** — at least one link/attachment: error-tracker issue, log excerpt, stack trace, screenshot/recording.
+
+A Bug's terminal state is its reproduction: the same steps (or test) fail before the fix and pass after — capture both as evidence: under the S14 manifest when `runtime_behavior_change = true`, or attached directly to the item for non-runtime Bugs (doc/config fixes), where S14 is N/A.
+
+#### S6 — Spike-specific content
+
+When `issue_type = Spike`, description must include:
+
+- The **question** being answered
+- The **decision** the answer enables, and the options being weighed
+- A **timebox**
+- **Deliverable format and location** — decision doc / prototype / findings page, and where it will live, so the terminal state — a deliverable at that location that actually answers the question, with findings, decision and options — is checkable
+
+Gherkin AC is intentionally N/A for Spikes (S4).
+
+#### S7 — Project parent declared
+
+When `issue_type ∈ {Story, Task, Improvement}` AND the item is part of an Epic context, `parent_project_id` must be set. When `issue_type = Sub-task`, `parent_issue_id` must be set. (Validity of the IDs is checked in feasibility gates.)
+
+For top-level Bug / Spike not under an Epic, this gate is N/A.
+
+A **build-ready leaf work unit** that is not part of an Epic context stands alone: a flat `Task` / `Improvement`, or a childless `Story` / `Spike` (no open child work) with `build_ready = true`, is an independently claimable leaf per `leaf-only-lifecycle` ("must not be stranded"), so a missing `parent_project_id` is `N/A`, not a FAIL. A `Sub-task` is exempt from this exception — it always requires `parent_issue_id`. This mirrors the leaf carve-out already in S10/S15.
+
+#### S8 — Target Backend Environment
+
+Every leaf work unit carries `## Target Backend Environment` in the description. The section is where `runtime_behavior_change` is persisted, so it is rendered for exempt work too — see the declaration grammar at the end of this gate. Read accepted environments from the exact configured keys of `.lisa.config.json` `deploy.branches`, never from a hardcoded list. Accept a human-confirmed bare exact configured key or `Confirmed: <env>`, automated `Inferred: <env> — evidence: <title|body|reproduction|hostname>`, automated `Assumption: <env> — remote default branch <branch>` for a unique reverse-map, or `Assumption: remote default branch <branch>` when no unique reverse-map exists. Human confirmation replaces an automated annotation with the bare key or `Confirmed: <env>`. For legacy bare values, use managed draft markers and current ticket content only; provider edit history is not required. A marker proves automation and requires re-annotation; otherwise unknown provenance plus conflicting evidence fails for confirmation. Validate the annotation shape/source and remote-default branch; validate `<env>` as an exact configured key whenever present. A valid branch-only assumption must not fail solely because its reverse-map is absent or ambiguous. Normalize built-in `prod` ↔ `production` only when exactly one of those keys is configured. No other aliases are valid.
+
+**The section persists `runtime_behavior_change`, so it is never simply skipped.** Work that changes no runtime behavior declares the exemption in place of an environment: `None — no runtime behavior change: doc-only` (or `config-only` / `type-only`). A container (a Project, or any Issue holding child work) declares `None — container: state rolls up from children`. The `None —` prefix is the machine discriminator; the words after it are for the human reading the Issue.
+
+Derive the flag from the stored content, never from a caller's word about a live Issue: an exact configured environment key means `true`, a `None —` declaration means `false`, and an **absent section means underivable** — never `false`. On a live Issue the stored declaration is authoritative, so a caller asserting a `runtime_behavior_change` that contradicts it **FAILs** this gate naming both values rather than silently preferring either; one of the two is wrong, and picking one quietly is how an unauditable gate gets built. A **proposed spec** with no section **FAILs** — `lisa-linear-write-issue` must render it before the write. A **live legacy Issue** with no section is `N/A` with a repair note routed to claim time: the same asymmetry S19 uses, for the same reason, because failing every existing Issue would turn a legacy queue red for a section no human had a way to add.
+
+#### S9 — Sign-in Required
+
+When `authenticated_surface = true`, description must contain `## Sign-in Required` naming the account/role and credential source.
+
+If the spec doesn't set `authenticated_surface`, infer it: scan the description and AC for sign-in / login / "as a {role} user" / authenticated route signals. If signals present and no `Sign-in Required` section: FAIL.
+
+#### S10 — Repository section, single-repo scope
+
+When `issue_type ∈ {Bug, Task, Sub-task, Improvement}` — or a **build-ready childless Story/Spike** (a claimable leaf per `leaf-only-lifecycle`) — description must contain `## Repository` naming exactly one repo. Multiple repos OR cross-repo references in AC: FAIL with recommendation `"Split into per-repo work units under a shared parent Story (see lisa-task-decomposition step 1.5)"`.
+
+An **Epic** (a Linear Project), or a **Story/Spike that still holds child work** (or is not build-ready): skipped (may span repos — coordination containers, not claimable leaf work units).
+
+This gate is `product_relevant: false` because cross-repo work units are not a product question — they are a decomposition error. Callers (`lisa-linear-to-tracker`, `lisa-notion-to-tracker`, etc.) MUST pre-split cross-repo work into per-repo work units during the decomposition phase per `lisa-task-decomposition` step 1.5; an S10 failure here indicates the agent skipped that step and must auto-split + revalidate before writing, not surface a clarifying comment to product.
+
+#### S11 — Validation Journey present
+
+When `runtime_behavior_change = true`, description must contain `## Validation Journey`. Skipped for doc-only / config-only / type-only / Epic. When `runtime_behavior_change` is **underivable** — a live legacy Issue whose description carries no Target Backend Environment declaration (S8) — this gate is `N/A` with the same repair note rather than `PASS`, because a requirement conditioned on a flag nobody recorded cannot be asserted either way.
+
+The caller controls strictness via `journey_followup: "auto"` or `"none"`:
+- `auto` (default): missing section returns `FAIL` with remediation `"Invoke lisa-linear-add-journey to append the section after create"`. The write path auto-fixes; dry-run path leaves it as a FAIL the caller must address.
+- `none`: missing section is a `FAIL` the caller will not auto-fix.
+
+#### S12 — Source Precedence (when artifacts attached)
+
+When `artifacts_attached = true`, description must include source-precedence guidance covering: business rules → PRD body, visual treatment → mocks, flow → prototypes, API/data → data artifacts. Cross-axis conflicts surfaced under `## Open Questions`.
+
+Accept either placement:
+- A dedicated `## Source Precedence` subsection, OR
+- A "Source Precedence" / "authoritative source" paragraph under `Technical Approach` covering the four axes.
+
+Detect by scanning for the phrase `Source Precedence` (case-insensitive) anywhere in the description AND verifying the four axes are each named. Missing the phrase OR any axis: FAIL with remediation naming the missing axes.
+
+If the spec doesn't set `artifacts_attached`, infer it the same way S9 infers sign-in: scan the description for design/mock/prototype/data-artifact references (design-tool links, "mock", "prototype", spreadsheet or API artifacts). If such artifacts are referenced and no source-precedence guidance exists: FAIL.
+
+#### S13 — Relationship Search documented
+
+The item must EITHER have at least one entry in `relations`, OR the description / a comment must contain a `## Relationship Search` block listing the git history queries and Linear MCP queries that were run with their outcomes.
+
+An item with zero relations and no documented search: FAIL.
+
+#### S14 — Evidence manifest binding (leaf work units)
+
+When `issue_type ∈ {Bug, Task, Sub-task, Improvement}` AND `runtime_behavior_change = true`, the `## Validation Journey` must declare at least one **typed** `[EVIDENCE: <artifact-type>: <name>]` marker. These markers are the work unit's **evidence manifest** — the exact, enumerated set of artifacts that must be captured and attached before the item may be closed (see the "Per-Work-Unit Evidence Contract" section of the `verification` rule, the Definition of Done in `verification-lifecycle`, and the evidence-manifest gate in `tracker-evidence`).
+
+Each marker must satisfy ALL of:
+
+- `<artifact-type>` is one of the fixed taxonomy: `screenshot`, `recording`, `http-transcript`, `cli-output`, `log-snippet`, `db-query-output`, `perf-trace`, `test-run-log`, `deploy-log`, `state-dump`. (The legacy `[SCREENSHOT: name]` form is accepted as `screenshot`.)
+- `<name>` is kebab-case and unique within the item.
+
+**A marker names an artifact, not an assertion.** An untyped marker (`[EVIDENCE: load-failure-handled-gracefully]`) is an assertion label with nothing to capture and must FAIL, with a remediation that shows the typed transformation (e.g. → `[EVIDENCE: screenshot: load-failure-error-state]`, `[EVIDENCE: perf-trace: pipeline-load-tti]`).
+
+FAIL when the Validation Journey is present but declares zero binding `[EVIDENCE: ...]` markers, when any binding marker is untyped or uses a type outside the taxonomy, or when any binding name is empty, duplicated, or not kebab-case. A behavior-changing work unit SHOULD declare both a success marker and an error/edge marker; a journey with only one binding marker passes but the remediation should recommend adding the error/edge case.
+
+Parse claiming markers by the exact `[EVIDENCE:` prefix. A cross-work-item pointer in the canonical form `[EVIDENCE-REF: <work-item-ref> | <artifact-type>: <kebab-case-name>]` is non-claiming. The Lisa 2.223.0 form `[EVIDENCE-REF: <tracker-ref>: <artifact-type>: <kebab-case-name>]` is also accepted as a legacy non-claiming alias; parse it from the right so the final two fields are type/name and a tracker URL may contain `:`. Exclude both forms from the manifest, S14's minimum-marker count, local marker type/name validation, and duplicate-name checks. Independently validate every `EVIDENCE-REF`: the native work-item reference must be non-empty and unambiguous, the artifact type must use the fixed taxonomy, and the name must be non-empty kebab-case. A malformed reference FAILs S14 as an invalid pointer but never becomes a local evidence obligation. A valid canonical or legacy reference may point to a sibling's artifact, but it never satisfies S14 for this item. Therefore a runtime-changing leaf whose journey contains only `EVIDENCE-REF` entries FAILs S14 for zero local claiming markers, not because a valid legacy reference is malformed. Quoting or code-formatting another item's `[EVIDENCE: ...]` marker does not make it a reference; writers must convert it to the canonical pipe form.
+
+This gate depends on S11. It is `N/A` for containers — a **Project** (the Epic equivalent), or any item with open child work (coordination containers, not work units) — and for leaf units with `runtime_behavior_change = false` (doc-only / config-only / type-only). If S11 fails because the Validation Journey is absent, S14 also FAILs (there is no manifest to bind) with remediation pointing back to `lisa-linear-add-journey`. It is likewise `N/A` with a repair note when `runtime_behavior_change` is **underivable** under S8 — reading an absent declaration as `false` is exactly the silent assumption that made this gate unauditable on a live Issue.
+
+#### S15 — Leaf-only build-ready
+
+Enforces the build-side of the vendor-neutral `leaf-only-lifecycle` rule: **only a leaf work unit may carry the build-ready role.** This is the symmetric write-side guard for the Linear validator — a container sitting in the configured `ready` state is a lifecycle error and must FAIL here, regardless of how the item was produced. (Mirrors the "build-ready is leaf-only" rule that `lisa-linear-write-issue` applies at write time.)
+
+**When the gate applies.** Run S15 whenever the item is build-ready — i.e. `build_ready = true`, or the item's live state is the configured `ready` state. If the item is not build-ready, S15 is `N/A` (nothing claims a non-ready item, so the invariant is vacuous).
+
+**Resolve container vs. leaf — structural first, then nominal.** Per `leaf-only-lifecycle` the classification is structural: an item is a **container** if it has child work, whatever its declared type; otherwise the **issue type** decides. Determine child work from (in order) `child_refs`, native sub-issues, project-member issues (an Epic is modeled as a Linear Project), and `blocked_by` / parent references — the same hierarchy resolution `lisa-linear-read-issue` uses. When validating a live identifier, query sub-issues / project members alongside the item fetch.
+
+Apply this decision and FAIL the two invariant-violating cases:
+
+1. **Container with child work + build-ready** — child work is present (any type that has open children), AND build-ready. FAIL. A parent organizes work; it is never claimed and implemented directly. Its lifecycle state rolls up from its children.
+2. **Childless Epic/Project + build-ready** — `issue_type = Epic` (a Linear Project) with **no** child work, AND build-ready. Still FAIL: an Epic/Project is a pure rollup container by design, and a childless one is an incomplete decomposition or a mis-applied role, not an implementable unit. (A childless Story or Spike is **not** failed here — the childless-parent exception in `leaf-only-lifecycle` promotes every childless non-Epic type to a build-ready leaf.)
+
+PASS (the childless-parent exception) when the item is build-ready and is a **leaf work unit**: it has **no** open child work and `issue_type ≠ Epic` (i.e. `Bug, Task, Sub-task, Improvement`, or a childless `Story` / `Spike`). A flat Task/Bug, or a childless Story/Spike with no sub-issues, is a valid build-ready leaf and must not be stranded.
+
+| issue_type | has child work | build-ready | S15 |
+|---|---|---|---|
+| Bug / Task / Sub-task / Improvement / Story / Spike | no | yes | **PASS** (leaf) |
+| any type | yes | yes | **FAIL** (structurally a container) |
+| Epic (Project) | no | yes | **FAIL** (childless Epic/Project — pure rollup container, exception does not apply) |
+| any | any | no | **N/A** (not build-ready) |
+
+Remediation: `"Build-ready is leaf-only per leaf-only-lifecycle. Move the build-ready role off this container onto its leaf children (or, for a childless Epic, decompose it into leaf children or reclassify it to a leaf type); a parent's lifecycle state rolls up from its children and is never set to ready directly."`
+
+`product_relevant: false` — a build-ready container is a lifecycle/decomposition error for the caller to repair, not a product question.
+
+#### S16 — Source Requirement traceability (PRD-sourced Issues)
+
+Answers "why was this done?": every Issue generated from a PRD must carry
+the requirement it exists to satisfy, quoted verbatim, at every level of
+the hierarchy — sub-issues included, so a leaf claimed by build-intake in
+isolation is self-explanatory.
+
+**When the gate applies.** Run S16 whenever the spec declares `prd_source`
+(all `*-to-tracker` decomposition paths set it). Without `prd_source`
+(ad-hoc issues with no PRD lineage) the gate is `N/A` — but if a
+`Source Requirement` section is present anyway, still validate its shape
+so a malformed section never passes silently.
+
+**What must be present.** a `Source Requirement` section (`##` markdown heading) containing:
+
+1. A link to the source PRD (the `**PRD**:` line), and
+2. At least one `**Requirement` line with **verbatim quoted text**, or the
+   explicit derived-work form (`Derived work supporting R3, R7 — no single
+   PRD section.`).
+
+Missing section, missing PRD link, empty/paraphrased requirement text
+(quotes shorter than a few words, or prose with no quotation), or a bare
+R-id with no quote: FAIL with remediation
+`"Add a Source Requirement section citing the PRD link and quoting the requirement(s) this issue satisfies verbatim (see the Source Requirement shared format in the *-to-tracker skills). Derived work must name the requirements it supports."`
+
+`product_relevant: true` — a issue whose requirement cannot be traced is
+a product-clarity problem: nobody can tell why the work exists.
+
+#### S17 — Improvement measurability
+
+When `issue_type = Improvement`, the description must define the improvement as a measured delta per `work-item-definition-of-ready`:
+
+- **Metric + measurement method** the agent can run (command, query, dashboard export)
+- **Baseline** — the current measured value (a number, not an adjective)
+- **Target** — the numeric value or bound that defines done
+
+Without a baseline and a target an Improvement has no verifiable terminal state and can never be autonomously closed. FAIL names the missing pieces; when no baseline exists yet, the remediation is to file measuring it as the first step. `N/A` for every other type.
+
+#### S18 — Stateless-pickup dry-run
+
+The autonomy gate, run last, on every build-ready leaf (use the S15 classification; `N/A` for containers and non-build-ready items). Simulate a stateless agent reading only this item and the links it can resolve — session knowledge about the codebase does not count, because the next claimant will not have it. List every question that agent would have to ask a human before starting work, before choosing between materially different implementations, or before declaring the work done.
+
+Zero questions → PASS. Any question → FAIL, with each question listed verbatim as its own remediation line — these are exactly the clarifying comments the caller posts to the source. The structure gates are proxies; this gate checks the readiness property itself: `ready` means a stateless agent can drive this item to its terminal state with zero human clarification (see `work-item-definition-of-ready`).
+
+#### S19 — Branch Plan derivation
+
+Enforces the `derived-branch-plan` rule: the `## Branch Plan` section is **derived only** — recompute it from current config and compare; never trust the rendered branches or read them as input. Resolve the environment under the S8 grammar, map that exact configured key forward through `.lisa.config.json` `deploy.branches`, and require the mapped branch on the remote.
+
+| Case | Verdict |
+|---|---|
+| `runtime_behavior_change = false` (doc-only / config-only / type-only) or a Project/container, with no plan | `N/A` — absence is correct; never demand one |
+| Exempt work **carrying** a `## Branch Plan` | **FAIL** — hand-authored branches on work that declared no runtime target |
+| `runtime_behavior_change` **underivable** (live legacy item, no Target Backend Environment declaration) | `N/A` with a repair note — absence is never read as `false` |
+| Plan matches the recomputed plan | PASS |
+| Plan conflicts with the recomputed plan | **FAIL** |
+| Plan missing the `Derived from:` provenance line and disagreeing | **FAIL** — treated as hand-authored |
+| `Branch from` and `PR into` name two different branches | **FAIL** — malformed; they are the same branch by construction |
+| Derivation hits a stop condition (env absent from `deploy.branches`, ambiguous / non-unique mapping, branch missing on the remote) | **FAIL** — the same stop `lisa-implement` takes; never default to `main` or the remote default |
+| Proposed spec (pre-write) for applicable work with no plan | **FAIL** — `lisa-linear-write-issue` must render it before the write |
+| Live **legacy** Issue for applicable work with no plan | `N/A` with a repair note — routed to claim time, where `lisa-implement` writes the derived assumption as a comment and proceeds |
+
+Failing a proposed spec is free; failing every existing Issue would turn a legacy queue red for a section no human had a way to add, so legacy absence is repaired at claim time instead.
+
+FAIL names both plans (rendered and recomputed) and points the remediation at **the environment, never the branch** — e.g. `"Branch Plan conflicts with the environment mapping. Rendered 'Branch from: release/staging'; recomputed from Target Backend Environment 'production' via deploy.branches → 'main'. Correct the environment, not the branch — the branches are derived. Then re-render."` Never silently choose between the two plans.
+
+#### S20 — Named-control reachability
+
+Enforces the vendor-neutral `control-reachability` rule: **a work item that names an existing test as a red-before-green control must say what makes that test reach the code the change touches.** A control whose reachability cannot be stated is not a control — its green tells the implementer nothing, and the Issue's own stopping rule then instructs a revert of a correct fix.
+
+**When the gate applies.** Run S20 when the body — `Acceptance Criteria`, `Technical Approach`, or `Validation Journey` — names an **existing** test AND predicts a state change for it when the work lands. Signals: "must go red", "currently passes and must fail", "fails before the fix and passes after", "if it still passes the fix did nothing", or any equivalent stopping rule.
+
+**When it must stay silent.** `N/A` otherwise. A Issue that introduces a **new** test rather than pinning an existing one carries no reachability obligation and must never be reported as incomplete on this basis. This half is load-bearing: a gate that fires on every work item is one callers learn to route around.
+
+**What must be present.** One marker per named control, inside `## Validation Journey`:
+
+```text
+[CONTROL: <test-identifier> | reaches: <input-or-field>]
+```
+
+Parse by the exact `[CONTROL:` prefix and split on the single `|`.
+
+- `<test-identifier>` — non-empty; the test file path, the test name, or both. It must be enough to run the test.
+- `reaches:` — the literal key, then a non-empty description of the **input, field, fixture key, argument, or state** that carries execution into the changed code.
+
+FAIL when a named existing-test control has no marker, when a marker is malformed (missing the `|`, an empty identifier, a missing or empty `reaches:` half), or when the `reaches:` half restates the assertion instead of naming an input (`reaches: the fix`, `reaches: the changed code path`). Naming the code is not enough; the marker answers "what in this fixture gets execution there".
+
+Remediation: `"Name the input or field that makes <test> reach the code this work changes, as [CONTROL: <test> | reaches: <field>]. If the fixture does not reach it, the test cannot observe this change — extend the fixture or specify a new test instead."`
+
+`product_relevant: true` — an unfalsifiable stopping rule is a specification defect, and whoever wrote the Issue is who can repair it. `lisa-linear-write-issue` renders the marker; `control-reachability` carries the implementer-side counterpart, which forbids acting on the stopping rule before the cause of an unmoved control is established.
+
+### Feasibility Gates (require Linear lookups; skip in dry-run if requested)
+
+#### F1 — Issue type valid in team
+
+For Issue types: confirm the team supports the issue type via `lisa-linear-access operation: get-team`. Linear issue types are typically per-team; check that the requested type exists.
+
+For Epic (Project): confirm the team allows projects.
+
+#### F2 — Project parent exists and is in same team
+
+When `parent_project_id` is set: fetch via `lisa-linear-access operation: get-project`, confirm it exists and belongs to the configured team.
+
+When `parent_issue_id` is set (Sub-task case): fetch via `lisa-linear-access operation: get-issue`, confirm the issue is non-Sub-task and in the same team / project.
+
+#### F3 — Linked items exist
+
+For each entry in `relations`, call `lisa-linear-access operation: get-issue` to confirm the identifier resolves. Flag broken identifiers.
+
+#### F4 — Required labels exist (or can be created)
+
+For each label referenced (`status:*`, `component:<name>`, `prd-*`), confirm via `lisa-linear-access operation: list-issue-labels` (or `lisa-linear-access operation: list-project-labels` for Project labels) that it exists OR is creatable. Linear labels are team-scoped or workspace-scoped; flag if the requested scope is wrong.
+
+#### F5 — Required external access provable
+
+The factory-gate rule: an input must not enter the pipeline unless the current runtime can actually
+reach every external surface the work requires. Enumerate the surfaces this item depends on:
+
+- artifact links in the body (documents, designs, dashboards, spreadsheets, recordings),
+- systems named by the description, acceptance criteria, or Validation Journey ("read the CloudWatch
+  alarms", "pull the copy from the Google Doc", "check the Sentry issues"),
+- tooling the work plainly implies (a deploy target, a database, a third-party API).
+
+For each surface, prove **read** access from the current runtime with a target-resource-specific,
+read-only probe through its sanctioned access layer: the matching MCP tool or `lisa-*-access` skill,
+or an authenticated fetch using environment-injected authentication. Identity-only commands such as
+`aws sts get-caller-identity`, `gh auth status`, and vendor equivalents are preflight checks only;
+they never satisfy F5 by themselves. A successful probe for one surface does not cover another:
+reading the named GitHub repository, CloudWatch log group, Sentry project, or linked document must
+each succeed separately when that surface is required.
+
+Attempt to resolve a gap before failing, but only through another configured brokered access layer
+or environment-injected authentication. A sanctioned broker may use its own credential store
+internally; the validator must never autonomously inspect, read, copy, print, or export raw
+credentials, keychains, credential files, or token stores, and must never invoke low-level secret
+tools to recover them. This preserves the intake agent's discover-first duty without exposing
+credential material.
+
+- `PASS` — every required surface has its own successful target-resource read probe or authenticated
+  fetch through the sanctioned access layer.
+- `N/A` — the item needs nothing beyond the repository and the tracker itself.
+- `FAIL` — a required surface is unreachable after the resolution attempt. Name the exact surface
+  and what was probed. Intake callers must route this to `blocked` + human escalation with the
+  missing access spelled out — an input the factory cannot execute never enters the factory.
+
+Probes are read-only and bounded (seconds, not minutes, per surface); never mutate the external
+system, and never invent or ask for credentials inline.
+
+## Execution
+
+1. Parse `$ARGUMENTS`. If it's an identifier, fetch the item and derive the spec from the fetched fields — including `runtime_behavior_change` (from the `## Target Backend Environment` declaration: an exact configured environment key → `true`, a `None —` declaration → `false`, an absent section → **underivable**, never `false`), `build_ready` (the Issue's state is the configured `ready` state) and `child_refs` (sub-issues, project-member issues, plus `blocked_by` parentage, resolved as in `lisa-linear-read-issue`) so S15 can classify the item. Otherwise parse the YAML spec.
+2. Resolve team ID via `lisa-linear-access operation: list-teams({query: <teamKey>})` if any feasibility gate will run.
+3. Run every Specification gate in order. Collect PASS / FAIL / N/A with a one-line reason.
+4. Unless the caller passed `--spec-only` (dry-run), run every Feasibility gate. Collect results.
+5. Emit the report below.
+
+## Output
+
+Output is a single fenced text block. Callers parse it; do not add free-form prose around it.
+
+```text
+## linear-validate-issue: <IDENTIFIER-or-SUMMARY>
+
+### Specification Gates
+- [PASS|FAIL|N/A] S1 Required core fields — <one-line reason>
+- [PASS|FAIL|N/A] S2 Summary format — <one-line reason>
+- [PASS|FAIL|N/A] S3 Description three audiences — <one-line reason>
+- [PASS|FAIL|N/A] S4 Acceptance criteria in Gherkin — <one-line reason>
+- [PASS|FAIL|N/A] S5 Bug-specific content — <one-line reason>
+- [PASS|FAIL|N/A] S6 Spike-specific content — <one-line reason>
+- [PASS|FAIL|N/A] S7 Project parent declared — <one-line reason>
+- [PASS|FAIL|N/A] S8 Target Backend Environment — <one-line reason>
+- [PASS|FAIL|N/A] S9 Sign-in Required — <one-line reason>
+- [PASS|FAIL|N/A] S10 Single-repo scope — <one-line reason>
+- [PASS|FAIL|N/A] S11 Validation Journey — <one-line reason>
+- [PASS|FAIL|N/A] S12 Source Precedence — <one-line reason>
+- [PASS|FAIL|N/A] S13 Relationship Search — <one-line reason>
+- [PASS|FAIL|N/A] S14 Evidence manifest binding — <one-line reason>
+- [PASS|FAIL|N/A] S15 Leaf-only build-ready — <one-line reason>
+- [PASS|FAIL|N/A] S16 Source Requirement traceability — <one-line reason>
+- [PASS|FAIL|N/A] S17 Improvement measurability — <one-line reason>
+- [PASS|FAIL|N/A] S18 Stateless-pickup dry-run — <one-line reason>
+- [PASS|FAIL|N/A] S19 Branch Plan derivation — <one-line reason>
+- [PASS|FAIL|N/A] S20 Named-control reachability — <one-line reason>
+
+### Feasibility Gates  (omit when --spec-only)
+- [PASS|FAIL|N/A] F1 Issue type valid in team — <one-line reason>
+- [PASS|FAIL|N/A] F2 Project parent exists and is in same team — <one-line reason>
+- [PASS|FAIL|N/A] F3 Linked items exist — <one-line reason>
+- [PASS|FAIL|N/A] F4 Required labels exist (or can be created) — <one-line reason>
+- [PASS|FAIL|N/A] F5 Required external access provable — <one-line reason>
+
+### Verdict: PASS | FAIL
+### Failures: <count>
+### Failure details
+- gate: <gate-id>
+  category: <category>
+  product_relevant: <true|false>
+  what: <plain-language description, no gate-IDs, no Linear jargon>
+  recommendation: <1–3 concrete options>
+- gate: ...
+```
+
+The verdict is `PASS` if and only if every applicable gate is `PASS`. Any `FAIL` makes the verdict `FAIL`. `N/A` does not affect the verdict.
+
+## Rules
+
+- Never write to Linear. The `allowed-tools` list intentionally excludes any `save_*` tool.
+- Never auto-fix the spec. This skill reports gaps; callers decide what to do.
+- Never silently skip a gate. If a gate genuinely doesn't apply, return `N/A` with the reason.
+- The `what` and `recommendation` fields must be concrete and product-readable — the dry-run path turns each failure into a Linear comment. Vague guidance is useless; always give 1–3 candidate resolutions.
+- Never emit a category outside the fixed set.
+- `product_relevant` is determined by the gate, not by the failure context.

package/plugins/lisa-devin/skills/lisa-linear-verify/SKILL.md

@@ -0,0 +1,65 @@
+---
+name: lisa-linear-verify
+description: "Verifies a Linear work item meets organizational standards by fetching the live item and running it through lisa-linear-validate-issue. Catches anything dropped or reformatted on write — same gates as the pre-write check, but applied to what Linear actually stored. Read-only."
+allowed-tools: ["Bash", "Skill"]
+---
+
+# Verify Linear Work Item: $ARGUMENTS
+
+Fetch the live Linear work item and run it through `lisa-linear-validate-issue`. This catches any field that was dropped or reformatted between the pre-write spec and what Linear stored.
+
+This skill is the destination of the `lisa-tracker-verify` shim when `tracker = "linear"`. Read-only — never writes.
+
+## Configuration
+
+Reads `linear.workspace`, `linear.teamKey` from `.lisa.config.json` (with `.local` override).
+
+## Input
+
+`$ARGUMENTS` is a Linear identifier:
+- An Issue identifier (e.g. `ENG-123`)
+- A Project URL or slug (e.g. `https://linear.app/<workspace>/project/<slug>-<id>`)
+
+If `$ARGUMENTS` is not parseable, stop and report.
+
+## Phase 1 — Resolve Context
+
+1. Read `linear.workspace`, `linear.teamKey` from `.lisa.config.json` (with `.local` override).
+2. Resolve team ID via `lisa-linear-access operation: list-teams({query: <teamKey>})`.
+3. Determine entity type from the identifier shape:
+   - `<TEAM>-<n>` → Issue
+   - URL containing `/project/<slug>-<id>` → Project
+
+## Phase 2 — Fetch Live State
+
+Call `lisa-linear-access operation: get-issue` (for Issues) or `lisa-linear-access operation: get-project` (for Projects). Capture every field, label, relation, comment, milestone, and project membership.
+
+## Phase 3 — Delegate to Validator
+
+Pass the fetched item to `lisa-linear-validate-issue` (in identifier mode — let it derive the spec from the live state). The validator runs both Specification AND Feasibility gates against what Linear actually stored.
+
+## Phase 4 — Report
+
+Return the validator's report verbatim — same structured format as `lisa-linear-validate-issue`. Callers (especially `lisa-linear-write-issue` Phase 7) parse the verdict to decide whether to declare success.
+
+If the verdict is `FAIL`, the caller should fix the item and re-run verify. Never declare success on a `FAIL` verdict.
+
+## Comparison semantics — semantic, never byte-exact
+
+Verification here re-runs `lisa-linear-validate-issue` against the live work item. It does **not** compare the
+stored body against the sent body byte for byte, and it must never drift to doing so.
+
+Linear normalizes markdown on write. Observed normalizations include rewriting `-` bullets
+as `*`, wrapping a bare URL as an explicit link, and re-segmenting bold emphasis around
+inline code spans — all lossless, all rendering-identical. **A byte-exact comparator cannot
+distinguish vendor markdown normalization from corruption**, so it reports failure on
+perfectly healthy writes and trains its reader to ignore it (CodySwannGT/lisa#3663).
+
+Any comparison of tracker-normalized rich text is therefore semantic, or
+normalize-then-compare. Byte-exact comparison of such text is forbidden.
+
+## Rules
+
+- Never write to Linear. Read-only.
+- Never short-circuit the validator. Always run the full gate set.
+- If `get_issue` / `get_project` returns an access error, surface it and exit — don't pretend the item is fine.