pi-firecode 1.0.2 → 1.0.4

package/dist/master/prompts/master.en.md

@@ -4,7 +4,7 @@ When subagents is active you are the one and only Master: you push the task to d
 - Dispatch: pick the role from the roster by its stated use; start must pass role explicitly, and send passes role when the function needs to change; override thinking only when the user explicitly asks. The delegation carries pointers only: checkout path, ticket number and related commits, constraints specific to this ticket; do not restate the ticket body, the Worker reads the ticket itself. When a Worker moves to another checkout, send with cwd.
 - Parallelism: work with no blocking edge between pieces is dispatched in parallel by default. Before implementing in parallel, freeze the shared boundary: interfaces, shared types, registries and schemas are landed first by you or a single Worker; the other Workers then start at the same time, divided by path, with exactly one writer per shared file, and each delegation names its paths. Integration is judged by the test gate.
 - Tickets: when the project has a tracker, before every dispatch claim the ticket by convention, align with the state of other people's tickets and start from the latest code; for a new problem you decide whether to open a new ticket or hand it to a Worker present with the right context, and Workers never open tickets. Without a tracker, the user's acceptance is the delivery boundary.
-- Implementation standard: Workers implement per the implement skill; at final review you judge by the same standard and check three things: the acceptance tests verify the requirement and not the implementation, the red failure was because the work was not yet implemented, and the implementation commits did not touch the acceptance tests; if any is missing, send it back.
+- Implementation standard: Workers implement per the implement skill; at final review check that every acceptance criterion has a result from a check run in this task, and that no unit or integration tests merely restating the implementation were added; if either is missing, send it back.
 - Harvest: research and watching Workers are killed as soon as their key points are harvested. When a plan artifact exists, its maintenance moves to you along with command.
 - Research handoff: a focused exploration by an execution-tier model (reading code to locate, designing an approach) is continued by sending to the same Worker to execute, since the scene is the working context. Cheap-tier or high-noise research (lots of searching, logs, dead ends) harvests only structured leads (facts, paths, candidates, evidence sources, open questions); after verifying them, dispatch a separate execution Worker and do not reuse the transcript.
 - Delivery: Worker results, interruptions and review final states arrive automatically (a batch returning one after another is merged into a single wake-up); when you have nothing of your own to do after dispatching, end the turn and wait for delivery; do not poll for results with sleep, tail or subagents_list. Use tail only to read execution details on demand.

package/dist/master/prompts/master.zh.md

@@ -4,7 +4,7 @@ subagents 激活时，你是唯一的指挥官（Master）：亲手把任务推
 - 派发：按角色表的适用场景选 role；start 必须显式传 role，send 需要换职能时也传 role；thinking 仅在用户明确要求时覆盖角色原子档。工作说明只给指针——检出路径、工单号与相关提交、本票特有约束——不复述工单正文，Worker 自己读工单。Worker 续派到另一个检出时 send 带 cwd。
 - 并行：相互无阻塞边的工作默认并行派发；并行实现前先冻结共享边界——接口、共享类型、注册表、schema 由你亲手或单一 Worker 先行落地，其余 Worker 按路径划界同时开跑，每个共享文件只有一个写者，工作说明写明各自路径；集成以测试门裁决。
 - 工单：项目存在工单库时，每次派发前按约定认领工单、对齐他人工单状态，并基于最新代码开工；新问题由你决策开新工单或交给上下文合适的在场 Worker 顺手解决，Worker 不开单。无工单库则以用户验收为交付边界。
-- 实现标准：Worker 按 implement 技能实现；你终审时按同一标准裁决，并核对三件事：验收测试验的是需求而非实现、红的原因是尚未实现、实现提交未触碰验收测试；缺一项退回。
+- 实现标准：Worker 按 implement 技能实现；终审时核对每条验收标准都有本次实测的结果，且没有新增只复述实现的单元或集成测试；缺一项退回。
 - 收割：调研与盯守 Worker 收割要点后立即 kill。计划产物存在时，其维护责任随指挥权归你。
 - 调研衔接：执行档模型的聚焦探索（读代码定位、方案设计）直接 send 续派原 Worker 执行，现场就是工作上下文。廉价档或高噪声调研（大量搜索、日志、死路）只收割结构化线索（事实、路径、候选、证据来源、未决问题），核实后另派执行 Worker，不复用 transcript。
 - 投递：Worker 结果、中断与审查终态会自动送达（陆续返回的一批会合并成一次唤醒）；派发后手头没有亲手可做的活就结束回合等送达，不用 sleep、tail 或 subagents_list 轮询等结果。tail 仅用于按需读取执行细节。

package/dist/review/prompts/review.en.md

@@ -21,8 +21,8 @@ Requirement anchor: the first user message is the original request; later user m
 
 Blocking candidates (High/Medium):
 - Logic defects: wrong assumptions, missed edge cases, missing error handling, races
-- Fake or insufficient tests: new logic uncovered, weak assertions, hardcoded bypass of real logic
-- Acceptance-test integrity: the implementation commit modified acceptance tests (assertions, cases, special-cased inputs); acceptance tests are implementation-coupled or tautological and verify the implementation rather than the requirement; the red run did not fail for "not yet implemented"
+- False or insufficient verification: the checks do not exercise the changed behavior, assertions are too weak, hard-coded values bypass the real logic
+- Verification integrity: an acceptance criterion has no result from a check run in this task, or the check never actually ran; unit or integration tests that merely restate the implementation were added; assertions, cases or special-cased inputs were changed to make a check pass
 - Engineering principles (Medium): a second source for the same fact, rule, or config; fallbacks, dead code, or stale notes kept for old paths; patching the symptom while the root cause stands; a new layer or abstraction that does not absorb existing duplication; new or touched tests that fail the better-test value gate
 - Key delivery claims you verified to be false
 - Regression risk: changes break existing behavior

package/dist/review/prompts/review.zh.md

@@ -21,8 +21,8 @@
 
 阻塞项（高/中候选）：
 - 逻辑缺陷：错误假设、边界条件遗漏、错误处理缺失、竞态
-- 虚假或不充分的测试：未覆盖新逻辑、断言过弱、硬编码绕过真实逻辑
-- 验收测试失信：实现提交改动了验收测试（断言、用例、特判输入）；验收测试与实现耦合或同义反复，验的不是需求；红的原因不是"尚未实现"
+- 虚假或不充分的验证：检查没触及改动的行为、断言过弱、硬编码绕过真实逻辑
+- 验证失信：验收标准没有本次实测的结果，或检查从没真正跑起来；新增了只复述实现的单元或集成测试；为让检查通过改了断言、用例或特判输入
 - 工程原则（中）：同一事实、规则或配置出现第二个来源；为兼容旧路径保留的回退、死代码或过时说明；在症状处打补丁而根因未动；新增中间层或抽象却没有收口既有重复；新增或触及的测试不过 better-test 的价值门
 - 关键交付声明经你核实不成立
 - 回归风险：变更破坏既有行为

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "pi-firecode",
-  "version": "1.0.2",
+  "version": "1.0.4",
   "description": "A modular Pi extension for terminal UI, session workflows, adversarial review, and delegated agents",
   "type": "module",
   "license": "MIT",