refacil-sdd-ai 5.0.1 → 5.0.2

package/README.md

@@ -6,6 +6,8 @@ Installs **skills** and **sub-agents** for **Claude Code**, **Cursor**, and **Op
 
 ---
 
+**npm:** [refacil-sdd-ai](https://www.npmjs.com/package/refacil-sdd-ai) | **GitHub:** [Erikole21/refacil-sdd-ai](https://github.com/Erikole21/refacil-sdd-ai)
+
 ## Requirements
 
 - **Node.js >= 20.0.0**

package/agents/debugger.md

@@ -48,6 +48,7 @@ If you prefer to continue here, provide:
 - **In mode=fix: you have Edit and Write** to implement the fix, generate tests, and create `summary.md`.
 - **The fix must be MINIMAL** — do not refactor anything beyond the bug.
 - **Return ONE final message** with the report + JSON block corresponding to the mode.
+- **Language policy for written files**: any source/tests/comments/identifiers and file/folder names written in mode=fix must be English-only. Only user-facing narrative may follow the user's language.
 
 ---
 

package/agents/implementer.md

@@ -59,6 +59,7 @@ Apply these 4 rules in each implementation task:
 - **Do NOT generate SDD artifacts** (proposal, specs, design, tasks) — that is `/refacil:propose`'s responsibility.
 - **Do NOT change branches or make commits** — the skill wrapper handles that before invoking you.
 - **Return ONE final message** with the report + JSON block.
+- **Language policy for implementation output**: all created/modified code artifacts must be in English (file/folder names, identifiers, test descriptions, and code comments), regardless of user language or SDD artifact language.
 
 ## Flow
 

package/agents/proposer.md

@@ -48,6 +48,7 @@ Exploration is necessary in this agent but must be **directed**, not exhaustive.
 - **NEVER write, modify, or generate source code** — only planning artifacts: `proposal.md`, `design.md`, `tasks.md`, specifications in `specs.md` and/or `specs/**/*.md`.
 - **Return ONE final message** with the summary + JSON block.
 - Your session context is isolated: explore with focus — depth in relevant modules, not breadth across the whole codebase.
+- **Language boundary**: `artifactLanguage` applies only to SDD artifact prose. If you include example identifiers/paths/snippets in artifacts, keep them in English and never generate real source code.
 
 ## Artifact templates
 

package/agents/tester.md

@@ -46,6 +46,7 @@ If you prefer to continue here, provide:
 - **You do NOT modify source code** — only generate test files.
 - **You do NOT create SDD planning artifacts** (proposal/specs/design/tasks) — that is `/refacil:propose`'s responsibility.
 - **Return ONE final message** with the report + JSON block.
+- **Language policy for tests**: generated test files must be English-only (file names, test names/descriptions, identifiers, and comments), regardless of user language or SDD artifact language.
 
 ## Stack detection (minimum focus)
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "refacil-sdd-ai",
-  "version": "5.0.1",
+  "version": "5.0.2",
   "description": "SDD-AI: Specification-Driven Development with AI — development methodology using AI with Claude Code, Cursor and OpenCode",
   "bin": {
     "refacil-sdd-ai": "./bin/cli.js"

package/skills/apply/SKILL.md

@@ -182,4 +182,5 @@ If the sub-agent returned `result: "PARTIAL"` or `"FAILED"`, present the `issues
 - Meet the preconditions of Step 0 (complete artifacts) and Step 1 (valid working branch) **before delegating**.
 - **Always build the briefing (Step 1.5) before delegating** — it is the key piece that reduces the sub-agent cost.
 - **Always delegate implementation to the sub-agent**. Do not replicate implementation logic or SDD artifact-apply logic here.
+- Even when artifacts are Spanish, implementation output is English-only: created file/folder names, source code, test code, identifiers, and code comments.
 - **Flow continuity**: if the user confirms affirmatively ("yes", "ok", "go", "continue", etc.) the continuity question in Step 3, immediately invoke the **Skill tool** with `skill: "refacil:test"`. Do not describe it in text or wait for the user to type `/refacil:test`. (See `METHODOLOGY-CONTRACT.md §5`.)

package/skills/bug/SKILL.md

@@ -165,5 +165,6 @@ If the sub-agent returned `result: "FAILED"` (tests not passing), present the fa
 - **NEVER implement without explicit user approval** (Step 3).
 - **Always validate the branch** (Step 4) before delegating the fix.
 - **Do not replicate investigation or implementation logic here** — that lives in `refacil-debugger`.
+- Bugfix implementation output is English-only: source/test file names, identifiers, test descriptions, and code comments, regardless of the language used in user-facing text or SDD artifacts.
 - Step 3 (bus cross-repo) is **optional** — only applies if the sub-agent reported `crossRepo: true`.
 - **Flow continuity**: if the user confirms affirmatively ("yes", "ok", "go", "continue", etc.) the continuity question in Step 6, immediately invoke the **Skill tool** with `skill: "refacil:review"`. Do not describe it in text or wait for the user to type `/refacil:review`. (See `METHODOLOGY-CONTRACT.md §5`.)

package/skills/prereqs/METHODOLOGY-CONTRACT.md

@@ -152,4 +152,9 @@ If the user does not request detail, use concise mode.
 - **Agent and skill internal instructions**: always in **English** (reduces token cost, improves LLM performance).
 - **Responses to the user**: in the **user's language**. If the user writes in Spanish, respond in Spanish. If in English, respond in English. Default: Spanish.
 - **SDD artifacts** (proposal.md, specs, design.md, tasks.md): in the **user's language** (or the language the team agreed on for the project).
-- Technical terms, code identifiers, and proper nouns stay in their canonical language regardless.
+- **Source code and generated files are always English-only**, regardless of `artifactLanguage`:
+  - code identifiers (variables, functions, classes, types, interfaces, enums)
+  - test code and test names/descriptions
+  - source file and folder names created during implementation
+  - code comments and commit/PR technical text produced by the agent
+- Never translate existing canonical API names, library symbols, or protocol/domain terms.

package/skills/propose/SKILL.md

@@ -105,4 +105,5 @@ Do you want me to continue with /refacil:apply?
 
 - **Change folder name**: always validate with `refacil-sdd-ai sdd validate-name <slug>` before delegating. Do not proceed if it exits 1.
 - **Always delegate generation to the sub-agent**. Do not replicate the codebase exploration or artifact generation logic here.
+- `artifactLanguage` affects **only SDD artifacts**. Any code snippets, file/folder names, identifiers, and technical comments that may appear during proposal work must stay in **English**.
 - **Flow continuity**: if the user confirms affirmatively ("yes", "ok", "go", "continue", etc.) the continuity question in Step 4, immediately invoke the **Skill tool** with `skill: "refacil:apply"`. Do not describe it in text or wait for the user to type `/refacil:apply`. (See `METHODOLOGY-CONTRACT.md §5`.)

package/skills/test/SKILL.md

@@ -100,4 +100,5 @@ Do you want me to continue with /refacil:verify?
 - **Always build the briefing in change mode (Step 1) before delegating** — reduces the sub-agent tool calls.
 - **Always delegate to the sub-agent**. Do not replicate stack detection or generation logic here.
 - **Do not invoke with ambiguous scope**. If there are multiple active changes, ask for selection first.
+- Test implementation is English-only (test file names, test cases/descriptions, identifiers, and comments), regardless of the SDD artifact language.
 - **Flow continuity**: if the user confirms affirmatively ("yes", "ok", "go", "continue", etc.) the continuity question **and tests passed (`passed: true`)**, immediately invoke the **Skill tool** with `skill: "refacil:verify"`. Do not describe it in text or wait for the user to type `/refacil:verify`. (See `METHODOLOGY-CONTRACT.md §5`.)