@pilotspace/add 1.11.0 → 1.13.0

package/tooling/add_engine/constants.py

@@ -213,7 +213,7 @@ _DELTA_RE = re.compile(
 )
 _EVIDENCE_RE = re.compile(r"^(.*?)\s*\(evidence:\s*(.*?)\)\s*$")
 _SPEC_DELTA_RE = re.compile(
-    r"\s*-\s*\[\s*(SPEC)\s*·\s*(open|seeded|dropped)\s*\]\s*(.+)$"
+    r"\s*-\s*\[\s*(SPEC)\s*·\s*(open|seeded|dropped|carried)\s*\]\s*(.+)$"
 )
 
 

package/tooling/add_engine/release.py

@@ -46,9 +46,9 @@ def _key_decisions_for(root: Path, slug: str) -> list[str]:
             if st.startswith(("-", "*")) and slug in st]
 
 def _build_in_flight(state: dict) -> bool:
-    """release_tests_red proxy (PURE): is any ACTIVE task mid-build without a recorded green gate
-    — phase ∈ {build, verify} AND gate == 'none'? The tool-agnostic engine never runs the suite,
-    so an entered-but-ungated build is the recorded-evidence stand-in for 'the suite is red'."""
+    """release_build_in_flight proxy (PURE): is any ACTIVE task mid-build without a recorded green
+    gate — phase ∈ {build, verify} AND gate == 'none'? The tool-agnostic engine never runs the
+    suite, so an entered-but-ungated build is the recorded-evidence stand-in for 'the suite is red'."""
     return any(t.get("phase") in ("build", "verify") and t.get("gate") == "none"
                for t in (state.get("tasks") or {}).values())
 

package/tooling/templates/DESIGN.md.tmpl

@@ -31,6 +31,17 @@ The design intent the AI drafts UI against — the persona, the one job, the fee
 - **Design principles** — <!-- 3–5 lines, e.g. "one primary action per screen" · "never hide state" -->
 - **Accessibility floor** — <!-- e.g. WCAG AA contrast · focus-visible · hit-target ≥ 44px -->
 
+## Design intake
+
+The per-design picks the `design-intake` beat collects (see the `add` skill's `design.md`).
+These are project **defaults**; a screen records only its **overrides** in its prototype note.
+VISUAL DESIGN draws its color/type from **## Identity** above — don't restate the brand here.
+
+- **Fidelity** — <!-- lo-fi wireframe | hi-fi mockup | production — how far this design renders -->
+- **Concept** — <!-- the design idea / mood / direction, in a line -->
+- **Layout** — <!-- the structure / grid / hierarchy the screens lean on -->
+- **Visual design** — <!-- spacing scale · imagery style (color · type come from ## Identity) -->
+
 ## Screens
 
 An index of the app's screens — one `design/prototypes/<name>.json` per screen (a flat

package/tooling/templates/TASK.fast.md.tmpl

@@ -55,6 +55,8 @@ Tests live in: `./tests/` · MUST run red (missing implementation) before Build.
 ## 5 · BUILD — AI writes code
 
 Scope (may touch): `./src/`   <every file the build may write — declared before the §3 freeze>
+Strategy & known-problem fixes: <ordered build steps · the trap each known problem must dodge>
+Strategy actually used: <fill at verify — what you ACTUALLY did, or "as planned"; harvested into §7 Decisions>
 Code lives in: `./src/`   ·   Constraints: change no test, no contract; allow-list packages only.
 
 ---

package/tooling/templates/TASK.md.tmpl

@@ -108,6 +108,8 @@ Tests live in: `./tests/` · MUST run red (missing implementation) before Build.
 
 Scope (may touch): `./src/`   <fill before the §3 freeze — every file the build may write>
 Strategy (ordered batches): <1. … 2. … — the planned build order; guidance, not enforced>
+Known-problem fixes: <trap → planned fix — the failure modes this build must dodge; guidance, not enforced>
+Strategy actually used: <fill at VERIFY — the strategy you ACTUALLY used (or "as planned"); harvested into the §7 Decisions (ADR) block as the [AI] build decision>
 Safety rule (feature-specific): <e.g. debit+credit in one atomic transaction>
 Code lives in: `./src/`
 Constraints: do NOT change any test or the contract; allow-list packages only; ask if unclear.
@@ -117,7 +119,8 @@ Constraints: do NOT change any test or the contract; allow-list packages only; a
      outside-root resolutions are dropped fail-closed · a DIRECTORY token covers its
      whole subtree (containment — diverges from §4's non-recursive counting) ·
      absent line = UNDECLARED (pre-existing tasks grandfathered, never retro-red) ·
-     engine enforcement (touched ⊆ declared) lands in scope-gate-enforce.
+     engine enforcement (touched ⊆ declared) is live: a completing verify gate refuses an
+     out-of-scope build (scope_violation → self-heal) and add.py check surfaces it.
      EXIT: all green; coverage held; no test/contract touched; no unlisted dependency. -->
 
 ---
@@ -145,6 +148,14 @@ Constraints: do NOT change any test or the contract; allow-list packages only; a
 - [ ] DEAD-CODE (code) — no new unused or orphaned symbol introduced
 - [ ] SEMANTIC (prose / non-code) — read in full, not skimmed: <what read · what confirmed>
 
+### Refute-read verdict — the earned-green check (record it; required for an auto-PASS)
+> Under autonomy: auto the AI auto-resolves Verify, so the earned-green refute-read MUST be
+> recorded here (the engine never spawns it — you do; NOT-EARNED -> `add.py heal`). The engine
+> MEASURES it is filled (`audit: refute_unrecorded`); it never auto-blocks — a human spot-audit
+> is the backstop. A human-gated (conservative/manual) task may leave it for the human's judgment.
+Verdict: <EARNED | NOT-EARNED>
+By: <self | agent-id> · adversarially checked: <what was probed>
+
 ### GATE RECORD
 Outcome: <PASS | RISK-ACCEPTED | HARD-STOP>
 If RISK-ACCEPTED -> owner: <name> · ticket: <link> · expires: <date>   (never for a security gap)
@@ -158,6 +169,9 @@ Reviewed by: <name> · date: <date>
 
 Watch (reuse scenarios as monitors): <error rate / per-rejection rate / latency>
 
+### Decisions (ADR)
+<harvested at done from §1/§3/§5/§6 — do not hand-edit; one actor-tagged line per decision, refilled only while this placeholder stands>
+
 ### Spec delta
 Forward changes for the next loop — each re-enters at Specify as the next task. One line
 each, tagged `[SPEC · open|seeded|dropped]`, with evidence (e.g. `[SPEC · open] rate-limit