bms-speckit-plugin 6.7.0 → 6.8.0

package/.claude-plugin/plugin.json

@@ -1,10 +1,10 @@
 {
   "name": "bms-speckit",
-  "version": "5.2.0",
-  "description": "Chain-orchestrated development pipeline with quality control agent. /bms-speckit runs brainstorm → constitution → specify → plan → tasks → analyze → implement → QC (UX/security/deps/code) → merge.",
+  "version": "6.8.0",
+  "description": "Chain-orchestrated development pipeline with 12-step autonomous workflow (brainstorm → research → constitution → specify → plan → tasks → analyze → implement with rolling QC → final quality gate → merge) and 9-dimension QC agent (code errors, security incl. SQL parameterization, deps, UX/UI, accessibility, deployment artifacts, cross-DB compatibility, real-DB integration testing, HOSxP business logic validation, brand/layout polish). Runs autonomously — never prompts for clarification.",
   "author": {
     "name": "manoirx"
   },
   "license": "MIT",
-  "keywords": ["workflow", "tdd", "specification", "planning", "development"]
+  "keywords": ["workflow", "tdd", "specification", "planning", "development", "chain", "orchestrator", "autonomous"]
 }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "bms-speckit-plugin",
-  "version": "6.7.0",
+  "version": "6.8.0",
   "description": "Chain-orchestrated development pipeline: /bms-speckit takes requirements and runs brainstorm → constitution → specify → plan → tasks → analyze → implement → verify with per-step error handling",
   "files": [
     ".claude-plugin/",

package/skills/bms-speckit-auto/SKILL.md

@@ -7,6 +7,20 @@ description: Full automated development workflow. This skill should be used when
 
 Chain blueprint: `blueprints/bms-speckit-pipeline.yaml`
 
+## Autonomous Mode Contract — READ FIRST
+
+This skill runs **fully autonomously**. Once invoked, it must not prompt the user for anything until the pipeline finishes or hits a hard error it cannot self-resolve.
+
+**Hard rules — no exceptions:**
+
+- **Never ask clarification questions.** Do not emit `[NEEDS CLARIFICATION: ...]` blocks, A/B/C option tables, or "what should I do?" prompts back to the user at any point in the pipeline, including inside Step 5 (specify).
+- **Resolve all unknowns with documented defaults.** If a requirement is ambiguous, pick the industry-standard default, record it in the spec's `## Assumptions` section with a one-line rationale, and continue. The user can correct it later — blocking the pipeline on a question costs more than picking wrong.
+- **Never request confirmation between steps.** No "proceed?", no "is this OK?". Commit and move on.
+- **Never stop for optional choices.** Only stop for hard failures (build broken, tests failing after 3 retries, critical security issue the QC agent flagged). Everything else: pick a default, log it, continue.
+- **Override sub-skill clarification behavior.** Skills like `speckit.specify` have their own "ask up to 3 clarification questions" logic. When invoked through this pipeline, that logic is **disabled** — the sub-skill must default the answers and record them as assumptions.
+
+**Wrong entry point check:** if the user invoked this pipeline via `/speckit.specify`, `/speckit.plan`, or any other individual speckit slash command instead of `/bms-speckit:bms-speckit-auto`, those individual commands run only their single step. They will not execute the full 12-step chain. Only this skill (`bms-speckit-auto`) orchestrates the complete pipeline.
+
 ## Execution Rules
 
 1. Execute steps **sequentially** in chain order — never skip steps
@@ -56,6 +70,8 @@ Subagent prompt:
 """
 You are running the BMS Speckit specification and planning chain. Execute each step in order using the Skill tool. Do NOT ask for confirmation. Follow error handling policies exactly. **Commit and push after every important step.**
 
+**AUTONOMOUS MODE — NO USER PROMPTS.** This subagent runs without user interaction. You must not return control to the user with questions, option tables, or `[NEEDS CLARIFICATION]` markers under any circumstance. For every ambiguity: pick the industry-standard default, record it in the relevant artifact's `## Assumptions` section with a one-line rationale, and continue. Specifically, when invoking `speckit.specify` in Step 5, pass it the explicit instruction: "Do not emit `[NEEDS CLARIFICATION]` markers or option tables; resolve all ambiguities with documented defaults in an `## Assumptions` section and produce a complete spec in one pass." Verify the final `spec.md` contains zero clarification markers before committing. If the sub-skill still emits markers, the subagent must rewrite the spec to replace each marker with a default + assumption before committing.
+
 **IMPORTANT — Progress reporting:** Before starting each step, output a progress message to the user in this format:
 
 `[Step N/12] step_name — description...`
@@ -117,7 +133,9 @@ After completing each step, output:
 - **Skill:** `speckit.specify`
 - **Input:** Use the detailed specification output from Step 1 (brainstorm) **enriched with the research findings from Step 2**. Include recommended libraries, best practices, and compliance requirements from the research brief in the specification input.
 - **Knowledge lookup:** Use `mcp__bms-knowledge-mcp__search_knowledge` to search the `hosxp` collection for exact table names, field names, data types, and relationships needed by this feature. Reference actual HOSxP data structures in the spec.
-- **Output:** `specs/*/spec.md`
+- **Autonomy override (CRITICAL):** `speckit.specify` by default emits `[NEEDS CLARIFICATION]` markers and prompts the user with A/B/C option tables for ambiguous requirements. **This behavior is DISABLED in the auto pipeline.** Pass this instruction to the specify skill invocation verbatim:
+  > "Do NOT emit any `[NEEDS CLARIFICATION]` markers or option tables. For every ambiguous requirement, pick the industry-standard default, write the chosen value into the spec directly, and record it in a new `## Assumptions` section at the end of the spec with a one-line rationale. The pipeline is running autonomously — the user will correct assumptions after the run if needed. Never block the pipeline with a question."
+- **Output:** `specs/*/spec.md` — must contain zero `[NEEDS CLARIFICATION]` markers. Verify before commit; if any remain, resolve them with a default and record in `## Assumptions`.
 - **Timeout:** 300s
 - **Retry:** up to 2 attempts
 - **Post-action:** Commit all files and push. Message: `feat(speckit): add feature specification`