@martinffx/atelier 0.1.0 → 1.0.1

package/CHANGELOG.md

@@ -1,5 +1,78 @@
 # Changelog
 
+## [1.0.1](https://github.com/martinffx/atelier/compare/v1.0.0...v1.0.1) (2026-05-23)
+
+
+### Bug Fixes
+
+* **ci:** resolve npm publish authentication failure ([7c57b85](https://github.com/martinffx/atelier/commit/7c57b8504e1f8045b9efeb7ccc5aebfb98e80050))
+* **ci:** resolve npm publish authentication failure in release workflow ([5e584bb](https://github.com/martinffx/atelier/commit/5e584bb7bea0f261152cc057dfab444e4c9c1402))
+
+## 1.0.0 (2026-05-22)
+
+
+### ⚠ BREAKING CHANGES
+
+* consolidate to command-based plugins
+
+### Features
+
+* add claude code atelier plugin marketplace ([06fcf59](https://github.com/martinffx/atelier/commit/06fcf594f4d9889c6ce0196aad468e9e72a931b1))
+* Add CLI to configure code harness ([#7](https://github.com/martinffx/atelier/issues/7)) ([64895ae](https://github.com/martinffx/atelier/commit/64895aec069fe13fb1ab739b5db8a4197f67b4e1))
+* add git workflow skills - worktree, graphite, and complete-branch ([22e2721](https://github.com/martinffx/atelier/commit/22e272129177e656f07dac76124d1e281f9ed4b4))
+* add parallel execution skill for dispatching parallel agents ([cde2e25](https://github.com/martinffx/atelier/commit/cde2e25a7f60d29fb42efc494546df3ea0706bde))
+* add planning skill for creating implementation plans with bite-sized tasks ([fc9dd3c](https://github.com/martinffx/atelier/commit/fc9dd3c033bfbec639ea8bdd6efbfd8fff6ec5f8))
+* add research skill for systematic exploration before planning ([d34bdbc](https://github.com/martinffx/atelier/commit/d34bdbcff25c5a93f3624a59f488f8d9d04300d0))
+* add spec:install skill and remove command shims ([7d6c693](https://github.com/martinffx/atelier/commit/7d6c693f195a570ace08477be68c9a2e22d7318b))
+* add using-atelier meta skill for skill invocation governance ([e35bee5](https://github.com/martinffx/atelier/commit/e35bee508b725b04a8b868d00c733ac879f73007))
+* add verification skill for evidence-based completion claims ([b2faaf2](https://github.com/martinffx/atelier/commit/b2faaf27dca2a3b1363bfdf4d3f7fb67059686b7))
+* **agents:** add oracle, architect, clerk agents and spec commands ([8eea909](https://github.com/martinffx/atelier/commit/8eea909100297dc4164b9c6a62e9f917a7ec0943))
+* **code:review:** add agent dispatch and skill loading ([f629fd9](https://github.com/martinffx/atelier/commit/f629fd92674ed2bc8f95d980d2eb518f3cdd6f92))
+* **code:review:** add gfreview integration with line-by-line comments ([586cf53](https://github.com/martinffx/atelier/commit/586cf53070e9dc7579d0e9db51c7028ce6652c7c))
+* enhance architect with planning patterns and execution handoff ([f80416a](https://github.com/martinffx/atelier/commit/f80416a0e80f6985d837ffb5149c8de3d0c189e7))
+* enhance debugging with systematic phases, iron law, and rationalization prevention ([01432c3](https://github.com/martinffx/atelier/commit/01432c3525761ff41a23a857ffbeba2605b01268))
+* enhance product discovery with brainstorming patterns and hard gates ([0dcaa39](https://github.com/martinffx/atelier/commit/0dcaa39bdb247c2cba469f2d30d455dc716dc6d5))
+* enhance testing with domain boundary TDD, iron law, and verification checklist ([5e8baa9](https://github.com/martinffx/atelier/commit/5e8baa98b70b7d12e9cb1f1fe4021d7fc63004b3))
+* **hooks:** add SessionStart hook to load spec:orchestrator ([6efcdf4](https://github.com/martinffx/atelier/commit/6efcdf41bf99862ee01eff7f320b3dfa5d7e2a38))
+* **plugin:** add code plugin with code-review and conventional-commit skills ([21ca1e3](https://github.com/martinffx/atelier/commit/21ca1e39f22cde25a9a6863f8fbcd43f7fe2462e))
+* **python:** add comprehensive Python ecosystem patterns plugin ([2a549c5](https://github.com/martinffx/atelier/commit/2a549c594f87643dac43b926db955f90aebfc7ed))
+* **skills:** add 5 new skills ([213aa61](https://github.com/martinffx/atelier/commit/213aa61bfe7e1259ee2e813b989f852b2757c2ee))
+* **skills:** add code-review and conventional-commit skills ([c0761b0](https://github.com/martinffx/atelier/commit/c0761b023160226e7d5a2fc468685903f6f51111))
+* **skills:** add product, architect, testing, and functional-patterns skills ([cd58be3](https://github.com/martinffx/atelier/commit/cd58be3a949533d7d8240a9d8f71bac84ab91b06))
+* **skills:** add spec:design skill for codebase research and solution design ([31d7980](https://github.com/martinffx/atelier/commit/31d7980789a314408676b2854e5c203d61a53473))
+* **skills:** add spec:implement skill for executing tasks from approved plan ([1318d9d](https://github.com/martinffx/atelier/commit/1318d9d72f424fbd91166d9a90c4529e334e925d))
+* **skills:** add spec:plan skill for implementation planning and task creation ([b0e51b5](https://github.com/martinffx/atelier/commit/b0e51b5e30b2466a1a3f26879b6deaee136a83e7))
+* **skills:** add spec:subagents skill for subagent dispatch patterns ([097d8cf](https://github.com/martinffx/atelier/commit/097d8cf2520ae6095dae5eb79fa8de5df9b5f2cf))
+* **skills:** add stacked-commit skill using Graphite ([a91ee82](https://github.com/martinffx/atelier/commit/a91ee82baa333bd013cef102153f0bd9de280afa))
+* **skills:** add using-atelier skill for skill invocation guidance ([58b0297](https://github.com/martinffx/atelier/commit/58b02970267848bdf36d8cc7d10dbff6c4e1c538))
+* **skills:** add using-git-worktrees skill ([55a31c9](https://github.com/martinffx/atelier/commit/55a31c903de752bedd1cf7c2bda2ed3ba396b5a1))
+* **spec:** add beads skill for task tracker CLI reference ([3ba4207](https://github.com/martinffx/atelier/commit/3ba4207c61a0afce4d54273300f1e17644c764d7))
+* **spec:** add dedicated /spec:init command with guided setup ([1fd074c](https://github.com/martinffx/atelier/commit/1fd074c10470db0705a28564df086c2911618651))
+* **spec:** add structured workflows with parallel execution and validation ([1671622](https://github.com/martinffx/atelier/commit/1671622b70c15c8a3a1abe662d82857847040a2e))
+* **spec:** separate workflow into distinct phase commands with validation ([24fb366](https://github.com/martinffx/atelier/commit/24fb366966118a25c30abc7a5d86b5044b530a60))
+* **typescript:** add build-tools skill with tooling references ([acb82e8](https://github.com/martinffx/atelier/commit/acb82e894d3e8ac834bf1612af000844222b9822))
+* **typescript:** add comprehensive testing skill for Vitest and MSW ([42c751f](https://github.com/martinffx/atelier/commit/42c751f4acbe93461ec457435b2181c57ee143f4))
+* **typescript:** add Effect-TS skill for functional effects ([#1](https://github.com/martinffx/atelier/issues/1)) ([558e81c](https://github.com/martinffx/atelier/commit/558e81c7cd3faf710206d7cdb4bdc7783e77f8c9))
+* **typescript:** add production patterns from exchequerio ([f217d20](https://github.com/martinffx/atelier/commit/f217d2027eb968ae5728fe64914d96b479dc7f28))
+* **typescript:** add sqlite, d1, and do to drizzle ([0886bb0](https://github.com/martinffx/atelier/commit/0886bb0525f34fa8ade24d9c1cd7b1b75ef03e9a))
+* **typescript:** enhance dynamodb-toolbox skill with production patterns ([e115371](https://github.com/martinffx/atelier/commit/e11537116d880cdfc83db15e267b77f8e554a7a5))
+* update methodology with superpowers integration, hard gates, verification, and domain boundary testing ([d68b133](https://github.com/martinffx/atelier/commit/d68b1332d97c4edde871ed91278a577f98c40a73))
+
+
+### Bug Fixes
+
+* **architect:** correct dependency inversion principle wording ([9c42952](https://github.com/martinffx/atelier/commit/9c429527cbf6d8b0567d9fc8f9bed635949bf76d))
+* **marketplace:** correct schema format for Claude Code validation ([5135e64](https://github.com/martinffx/atelier/commit/5135e64b56ba78574293ecca3c1669708149a028))
+* **skills:** update spec skills with embedded schemas and skill loading ([be76e7c](https://github.com/martinffx/atelier/commit/be76e7cd9ad132ba87c9dc3b419f2c88d5e44ded))
+* **skills:** update spec skills with skill loading and fix spec:subagents references ([498b3dc](https://github.com/martinffx/atelier/commit/498b3dc2ed91b5cd0e3ae9a55ef0d4e330da0422))
+* update skill names to match folder names ([649d1a4](https://github.com/martinffx/atelier/commit/649d1a4ca64a1691f407fe23f0f6d1550c10ac18))
+* **using-git-worktrees:** update integration section for atelier plugins ([e508f30](https://github.com/martinffx/atelier/commit/e508f30f38c40182c3781ab2b1113b5e97018ff4))
+
+
+### Code Refactoring
+
+* consolidate to command-based plugins ([8963fe4](https://github.com/martinffx/atelier/commit/8963fe4b6f189e67eae5eecccfeed6a6658ba33b))
+
 ## [0.1.0] - 2026-05-08
 
 ### Added

package/agents/scout.md

@@ -0,0 +1,40 @@
+---
+name: scout
+description: Fast codebase reconnaissance and exploration, concise summarization of findings, precise location of relevant context, and efficient workspace preparation through targeted file operations and template application
+model: minimax-m2.7
+---
+
+You are the **Scout**, a fast reconnaissance agent. Your job is to explore codebases, summarize what you find, highlight relevant context, and point others to exactly where things live. You move quickly, read minimally, and never break existing work.
+
+## Role
+
+- Explore and map codebase structure, conventions, and key files
+- Summarize findings with clear, concise overviews
+- Highlight relevant context and explain why it matters
+- Point to exact locations (file paths, line numbers, symbol names) so others can dive deeper
+- Retrieve targeted snippets on request—never dump full files unless necessary
+- Create files from templates and set up directory structures when asked
+- Respect existing files—never overwrite without explicit direction
+
+## Skills
+
+Before beginning work, scan your environment for relevant skills and load any that apply to the task at hand. Use them to guide how you explore, summarize, and present your findings.
+
+## Checklist
+
+Before finishing, confirm you have:
+
+- [ ] Retrieved only the specific context requested (no unnecessary full-file reads)
+- [ ] Verified the target file does not already exist before creating
+- [ ] Created parent directories as needed
+- [ ] Applied all relevant variable substitutions when using templates
+- [ ] Confirmed the output matches the request exactly
+- [ ] Reported what was created, retrieved, or modified
+
+## Boundaries
+
+- DO handle file operations and retrieval efficiently
+- DO create files and directories as requested
+- DON'T make design or architectural decisions 
+- DON'T conduct requirements discovery 
+- NEVER overwrite existing files unless explicitly told to

package/dist/atelier.js

@@ -25616,42 +25616,6 @@ function handleError(error) {
   process.exit(1);
 }
 
-// src/generators/directives.ts
-var ATELIER_DIRECTIVE = `
-MANDATORY SKILL ACTIVATION SEQUENCE \u2014 APPLIES TO EVERY INTERACTION
-
-BEFORE RESPONDING TO ANY PROMPT \u2014 Step 1 (EVALUATE):
-Check ALL skills in <available_skills>. State each as YES/NO with reason.
-
-BEFORE RESPONDING TO ANY PROMPT \u2014 Step 2 (ACTIVATE):
-For spec-orchestrator: ALWAYS YES. Invoke using the skill tool.
-For any other skill marked YES: Invoke using the skill tool.
-
-BEFORE RESPONDING TO ANY PROMPT \u2014 Step 3 (PROCEED):
-Only after Step 2 is complete, begin your response.
-
-BEFORE MAKING ANY CODE CHANGE:
-1. EVALUATE which language and framework skills in <available_skills> apply
-2. ACTIVATE all relevant language skills using the skill tool
-3. PROCEED only after skill loading
-
-BEFORE STARTING MULTI-STEP WORK:
-1. EVALUATE if subagents can parallelize the work
-2. DELEGATE to task tool with appropriate agent types
-3. PROCEED with main thread only after delegation is complete
-
-Preferred subagent types:
-- explore: Fast codebase discovery
-- recon: Targeted reconnaissance
-- general: Complex multi-step tasks
-- architect: System design decisions
-- oracle: Requirements discovery and critical thinking
-
-CRITICAL: Step 1 is WORTHLESS without Step 2. Use the skill tool to ACTIVATE.
-CRITICAL: Use subagents (task tool) for parallel work. Do NOT do everything in the main thread.
-THIS IS NON-NEGOTIABLE.
-`;
-
 // src/generators/opencode.ts
 var import_gray_matter2 = __toESM(require_gray_matter(), 1);
 var GLOBAL_OPENCODE_DIR = join2(homedir(), ".config", "opencode");
@@ -25740,11 +25704,14 @@ function deepMerge(target, source) {
   return result;
 }
 function writePluginJs(config, basePath) {
-  const plugin = `const ATELIER_CONTEXT = \`${ATELIER_DIRECTIVE}\`;
-
-export default {
+  const skillsPath = config.skills_path || "~/.agents/skills/atelier";
+  const skillsDir = `${skillsPath}/atelier`;
+  const plugin = `export default {
+  config: {
+    skillsDir: ${JSON.stringify(skillsDir)},
+  },
   'experimental.chat.system.transform': (systemPrompt) => {
-    return systemPrompt + '\\n\\n<!-- Atelier auto-injected context -->\\n' + ATELIER_CONTEXT;
+    return systemPrompt;
   },
 };
 `;
@@ -29920,7 +29887,6 @@ function generateClaude(config, basePath = process.cwd()) {
   try {
     writeSettingsJson(config, basePath);
     writeHookScript(config, basePath);
-    writeUserPromptHook(config, basePath);
     writeAgentFiles2(config, basePath);
   } catch (err) {
     throw new FileWriteError("generateClaude", err instanceof Error ? err.message : String(err));
@@ -29930,27 +29896,25 @@ function writeSettingsJson(config, basePath) {
   const settingsPath = join4(basePath, ".claude/settings.json");
   const existing = readExistingSettings(settingsPath);
   const isNew = Object.keys(existing).length === 0;
-  const sessionStartHook = {
-    hooks: [{ command: "hooks/atelier-session-start", type: "command" }],
+  const atelierHook = {
+    hooks: [
+      {
+        command: "hooks/atelier-session-start",
+        type: "command"
+      }
+    ],
     matcher: "startup|clear|compact"
   };
   const existingSessionStart = existing.hooks?.SessionStart || [];
-  const hasSessionStart = existingSessionStart.some((h) => h.hooks?.some((hook) => hook.command === "hooks/atelier-session-start"));
-  const newSessionStart = hasSessionStart ? existingSessionStart : [...existingSessionStart, sessionStartHook];
-  const userPromptHook = {
-    hooks: [{ command: "hooks/atelier-user-prompt", type: "command" }]
-  };
-  const existingUserPrompt = existing.hooks?.UserPromptSubmit || [];
-  const hasUserPrompt = existingUserPrompt.some((h) => h.hooks?.some((hook) => hook.command === "hooks/atelier-user-prompt"));
-  const newUserPrompt = hasUserPrompt ? existingUserPrompt : [...existingUserPrompt, userPromptHook];
+  const hasAtelierHook = existingSessionStart.some((h) => h.hooks?.some((hook) => hook.command === "hooks/atelier-session-start"));
+  const newSessionStart = hasAtelierHook ? existingSessionStart : [...existingSessionStart, atelierHook];
   const settings = {
     ...existing,
     $schema: "https://json.schemastore.org/claude-code-settings.json",
     model: config.default_model || "opusplan",
     hooks: {
       ...existing.hooks || {},
-      SessionStart: newSessionStart,
-      UserPromptSubmit: newUserPrompt
+      SessionStart: newSessionStart
     }
   };
   writeFileSync3(settingsPath, JSON.stringify(settings, null, 2) + `
@@ -29965,31 +29929,29 @@ function readExistingSettings(settingsPath) {
     return {};
   }
 }
+function escapeShellArg(arg) {
+  return `'${arg.replace(/'/g, "'\\''")}'`;
+}
 function writeHookScript(config, basePath) {
-  const additionalContext = JSON.stringify({ additionalContext: ATELIER_DIRECTIVE });
+  const skillsPath = config.skills_path || "~/.agents/skills/atelier";
+  const safeSkillsPath = escapeShellArg(skillsPath);
   const script = `#!/bin/bash
 # Atelier session-start hook
+# Skills path: ${skillsPath}
 
-echo '${additionalContext}'
+SKILLS_DIR=${safeSkillsPath}
+
+if [ -d "$SKILLS_DIR/atelier" ]; then
+  echo '{"additionalContext": {"skillsDir": "'"$SKILLS_DIR"'/atelier"}}'
+else
+  echo '{"additionalContext": {}}'
+fi
 `;
   const hookPath = join4(basePath, "hooks/atelier-session-start");
   writeFileSync3(hookPath, script);
   chmodSync(hookPath, 448);
   console.log("Created hooks/atelier-session-start");
 }
-function writeUserPromptHook(config, basePath) {
-  const script = `#!/bin/bash
-# Atelier user-prompt hook
-
-cat <<'EOF'
-${ATELIER_DIRECTIVE}
-EOF
-`;
-  const hookPath = join4(basePath, "hooks/atelier-user-prompt");
-  writeFileSync3(hookPath, script);
-  chmodSync(hookPath, 448);
-  console.log("Created hooks/atelier-user-prompt");
-}
 function writeAgentFiles2(config, basePath) {
   const agentsDir = join4(basePath, ".claude/agents");
   for (const agent of config.agents) {

package/package.json

@@ -1,13 +1,19 @@
 {
   "name": "@martinffx/atelier",
-  "version": "0.1.0",
+  "version": "1.0.1",
   "description": "Personal development toolkit - spec-driven development, code quality, deep thinking, and ecosystem patterns",
   "type": "module",
   "bin": {
     "atelier": "dist/atelier.js"
   },
   "repository": "github:martinffx/atelier",
-  "keywords": ["cli", "claude-code", "opencode", "ai-agents", "spec-driven-development"],
+  "keywords": [
+    "cli",
+    "claude-code",
+    "opencode",
+    "ai-agents",
+    "spec-driven-development"
+  ],
   "license": "MIT",
   "files": [
     "dist/",
@@ -17,6 +23,9 @@
     "LICENSE",
     "CHANGELOG.md"
   ],
+  "publishConfig": {
+    "access": "public"
+  },
   "scripts": {
     "build": "bun build ./src/atelier.ts --outdir ./dist --target bun",
     "typecheck": "tsgo --noEmit",

package/skills/spec-plan/SKILL.md

@@ -5,7 +5,7 @@ description: >
   when there's an approved spec (design.md) and the next step is breaking it into implementable
   work. Trigger when the user says "write a plan", "plan this out", "break this down",
   "I added notes" (annotation cycle), or after spec-brainstorm completes. Also trigger for
-  "create tasks". Do NOT use for research (use spec-brainstorm) or
+  "create tasks" or "add to beads". Do NOT use for research (use spec-brainstorm) or
   execution (use spec-implement).
 user-invocable: true
 ---