sam-agents 0.1.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (37) hide show
  1. package/LICENSE +21 -0
  2. package/README.md +92 -0
  3. package/bin/cli.js +320 -0
  4. package/package.json +36 -0
  5. package/templates/.claude/commands/sam/core/agents/sam.md +6 -0
  6. package/templates/.claude/commands/sam/core/workflows/autonomous-tdd.md +6 -0
  7. package/templates/.claude/commands/sam/sam/agents/argus.md +6 -0
  8. package/templates/.claude/commands/sam/sam/agents/atlas.md +6 -0
  9. package/templates/.claude/commands/sam/sam/agents/dyna.md +6 -0
  10. package/templates/.claude/commands/sam/sam/agents/iris.md +6 -0
  11. package/templates/.claude/commands/sam/sam/agents/sage.md +6 -0
  12. package/templates/.claude/commands/sam/sam/agents/titan.md +6 -0
  13. package/templates/_sam/_config/agent-manifest.csv +8 -0
  14. package/templates/_sam/_config/agents/sam-architect.customize.yaml +13 -0
  15. package/templates/_sam/_config/agents/sam-dev.customize.yaml +13 -0
  16. package/templates/_sam/_config/agents/sam-reviewer.customize.yaml +13 -0
  17. package/templates/_sam/_config/agents/sam-sam.customize.yaml +13 -0
  18. package/templates/_sam/_config/agents/sam-tech-writer.customize.yaml +13 -0
  19. package/templates/_sam/_config/agents/sam-test.customize.yaml +13 -0
  20. package/templates/_sam/_config/agents/sam-ux-designer.customize.yaml +13 -0
  21. package/templates/_sam/_config/ides/claude-code.yaml +6 -0
  22. package/templates/_sam/_config/manifest.yaml +13 -0
  23. package/templates/_sam/_config/workflow-manifest.csv +2 -0
  24. package/templates/_sam/agents/architect.md +69 -0
  25. package/templates/_sam/agents/dev.md +97 -0
  26. package/templates/_sam/agents/reviewer.md +138 -0
  27. package/templates/_sam/agents/tech-writer.md +147 -0
  28. package/templates/_sam/agents/test.md +120 -0
  29. package/templates/_sam/agents/ux-designer.md +141 -0
  30. package/templates/_sam/core/agents/sam-master.md +91 -0
  31. package/templates/_sam/core/config.yaml +31 -0
  32. package/templates/_sam/core/workflows/autonomous-tdd/steps/step-01-validate-prd.md +213 -0
  33. package/templates/_sam/core/workflows/autonomous-tdd/steps/step-02-generate-stories.md +266 -0
  34. package/templates/_sam/core/workflows/autonomous-tdd/steps/step-03-tdd-loop.md +422 -0
  35. package/templates/_sam/core/workflows/autonomous-tdd/steps/step-04-complete.md +356 -0
  36. package/templates/_sam/core/workflows/autonomous-tdd/templates/pipeline-status.yaml +108 -0
  37. package/templates/_sam/core/workflows/autonomous-tdd/workflow.md +311 -0
package/LICENSE ADDED
@@ -0,0 +1,21 @@
1
+ MIT License
2
+
3
+ Copyright (c) 2026 sam-agents
4
+
5
+ Permission is hereby granted, free of charge, to any person obtaining a copy
6
+ of this software and associated documentation files (the "Software"), to deal
7
+ in the Software without restriction, including without limitation the rights
8
+ to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
9
+ copies of the Software, and to permit persons to whom the Software is
10
+ furnished to do so, subject to the following conditions:
11
+
12
+ The above copyright notice and this permission notice shall be included in all
13
+ copies or substantial portions of the Software.
14
+
15
+ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16
+ IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17
+ FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18
+ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19
+ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20
+ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21
+ SOFTWARE.
package/README.md ADDED
@@ -0,0 +1,92 @@
1
+ # SAM - Smart Agent Manager
2
+
3
+ [![npm version](https://img.shields.io/npm/v/@sam-agents/sam.svg)](https://www.npmjs.com/package/@sam-agents/sam)
4
+ [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
5
+ [![Downloads](https://img.shields.io/npm/dm/@sam-agents/sam.svg)](https://www.npmjs.com/package/@sam-agents/sam)
6
+
7
+ **Autonomous TDD agent system for Claude Code and Cursor.**
8
+
9
+ SAM orchestrates a team of specialized AI agents that transform your PRD into working, tested code using strict Test-Driven Development (RED-GREEN-REFACTOR).
10
+
11
+ ## Quick Start
12
+
13
+ ```bash
14
+ # For Claude Code (default)
15
+ npx @sam-agents/sam
16
+
17
+ # For Cursor
18
+ npx @sam-agents/sam --platform cursor
19
+
20
+ # For all platforms
21
+ npx @sam-agents/sam --platform all
22
+ ```
23
+
24
+ ## Supported Platforms
25
+
26
+ | Platform | Install Command | Skill Format |
27
+ |----------|-----------------|--------------|
28
+ | **Claude Code** | `npx @sam-agents/sam` | `/sam:` commands |
29
+ | **Cursor** | `npx @sam-agents/sam --platform cursor` | `@agent` mentions |
30
+
31
+ ## Why SAM?
32
+
33
+ | Feature | SAM | Traditional AI Coding |
34
+ |---------|-----|----------------------|
35
+ | **TDD-First** | Tests written before code | Code first, tests maybe |
36
+ | **BYOA** | Use your own AI subscription | Pay per API call (5-10x cost) |
37
+ | **Transparency** | Watch agents work in real-time | Black box |
38
+ | **Autonomous** | Minimal intervention after PRD | Constant hand-holding |
39
+ | **Multi-Platform** | Claude Code + Cursor | Single platform lock-in |
40
+
41
+ ## Available Agents
42
+
43
+ | Agent | Claude Code | Cursor | Role |
44
+ |-------|-------------|--------|------|
45
+ | **SAM** | `/sam:core:agents:sam` | `@sam` | Orchestrator |
46
+ | **Atlas** | `/sam:sam:agents:atlas` | `@atlas` | System Architect |
47
+ | **Titan** | `/sam:sam:agents:titan` | `@titan` | Test Architect (RED) |
48
+ | **Dyna** | `/sam:sam:agents:dyna` | `@dyna` | Developer (GREEN) |
49
+ | **Argus** | `/sam:sam:agents:argus` | `@argus` | Code Reviewer (REFACTOR) |
50
+ | **Sage** | `/sam:sam:agents:sage` | `@sage` | Technical Writer |
51
+ | **Iris** | `/sam:sam:agents:iris` | `@iris` | UX Designer |
52
+
53
+ ## The TDD Pipeline
54
+
55
+ **Claude Code:** `/sam:core:workflows:autonomous-tdd`
56
+ **Cursor:** `@sam-tdd`
57
+
58
+ 1. **Validate PRD** - Atlas + Iris review requirements
59
+ 2. **Generate Stories** - Break down into epics and user stories
60
+ 3. **TDD Loop** - For each story:
61
+ - **RED**: Titan writes failing tests
62
+ - **GREEN**: Dyna writes minimal code to pass
63
+ - **REFACTOR**: Argus improves code quality
64
+ 4. **Complete** - Sage generates documentation
65
+
66
+ ## What Gets Installed
67
+
68
+ ```
69
+ your-project/
70
+ โ”œโ”€โ”€ _sam/ # Agent definitions (shared)
71
+ โ”‚ โ”œโ”€โ”€ agents/ # Individual agent configs
72
+ โ”‚ โ””โ”€โ”€ core/workflows/ # TDD pipeline workflow
73
+ โ”œโ”€โ”€ .claude/commands/sam/ # Claude Code skills (--platform claude)
74
+ โ””โ”€โ”€ .cursor/rules/ # Cursor rules (--platform cursor)
75
+ ```
76
+
77
+ ## Requirements
78
+
79
+ - Node.js 16+
80
+ - [Claude Code CLI](https://docs.anthropic.com/en/docs/claude-code) or [Cursor](https://cursor.com)
81
+
82
+ ## Contributing
83
+
84
+ We welcome contributions! See [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.
85
+
86
+ ## License
87
+
88
+ MIT - see [LICENSE](LICENSE) for details.
89
+
90
+ ---
91
+
92
+ Built with AI. Powered by TDD.
package/bin/cli.js ADDED
@@ -0,0 +1,320 @@
1
+ #!/usr/bin/env node
2
+
3
+ const fs = require('fs');
4
+ const path = require('path');
5
+ const readline = require('readline');
6
+
7
+ const CYAN = '\x1b[36m';
8
+ const GREEN = '\x1b[32m';
9
+ const YELLOW = '\x1b[33m';
10
+ const RED = '\x1b[31m';
11
+ const RESET = '\x1b[0m';
12
+ const BOLD = '\x1b[1m';
13
+ const DIM = '\x1b[2m';
14
+
15
+ const PLATFORMS = ['claude', 'cursor', 'all'];
16
+
17
+ function log(message, color = RESET) {
18
+ console.log(`${color}${message}${RESET}`);
19
+ }
20
+
21
+ function copyRecursive(src, dest) {
22
+ const stats = fs.statSync(src);
23
+
24
+ if (stats.isDirectory()) {
25
+ if (!fs.existsSync(dest)) {
26
+ fs.mkdirSync(dest, { recursive: true });
27
+ }
28
+
29
+ const files = fs.readdirSync(src);
30
+ for (const file of files) {
31
+ copyRecursive(path.join(src, file), path.join(dest, file));
32
+ }
33
+ } else {
34
+ const destDir = path.dirname(dest);
35
+ if (!fs.existsSync(destDir)) {
36
+ fs.mkdirSync(destDir, { recursive: true });
37
+ }
38
+ fs.copyFileSync(src, dest);
39
+ }
40
+ }
41
+
42
+ function countFiles(dir) {
43
+ let count = 0;
44
+ const items = fs.readdirSync(dir);
45
+ for (const item of items) {
46
+ const fullPath = path.join(dir, item);
47
+ if (fs.statSync(fullPath).isDirectory()) {
48
+ count += countFiles(fullPath);
49
+ } else {
50
+ count++;
51
+ }
52
+ }
53
+ return count;
54
+ }
55
+
56
+ function showHelp() {
57
+ log('\n' + BOLD + ' SAM - Smart Agent Manager' + RESET);
58
+ log(' Autonomous TDD Agent System\n', CYAN);
59
+ log(' Usage: npx @sam-agents/sam [options] [target-directory]\n');
60
+ log(' Options:');
61
+ log(' --platform <name> Target platform: claude, cursor, all');
62
+ log(' --help, -h Show this help message');
63
+ log(' --version, -v Show version number\n');
64
+ log(' Examples:');
65
+ log(' npx @sam-agents/sam Interactive mode');
66
+ log(' npx @sam-agents/sam --platform cursor Install for Cursor');
67
+ log(' npx @sam-agents/sam --platform all Install for all platforms');
68
+ log(' npx @sam-agents/sam ./myapp Install in ./myapp directory\n');
69
+ log(' Supported Platforms:');
70
+ log(' claude - Claude Code CLI (.claude/commands/)');
71
+ log(' cursor - Cursor IDE (.cursor/rules/)\n');
72
+ }
73
+
74
+ function generateCursorRules(samDir, targetDir) {
75
+ const cursorDir = path.join(targetDir, '.cursor', 'rules');
76
+
77
+ if (!fs.existsSync(cursorDir)) {
78
+ fs.mkdirSync(cursorDir, { recursive: true });
79
+ }
80
+
81
+ const agents = [
82
+ { name: 'sam', file: 'core/agents/sam-master.md', display: 'SAM Orchestrator' },
83
+ { name: 'atlas', file: 'agents/architect.md', display: 'Atlas - System Architect' },
84
+ { name: 'dyna', file: 'agents/dev.md', display: 'Dyna - Developer' },
85
+ { name: 'titan', file: 'agents/test.md', display: 'Titan - Test Architect' },
86
+ { name: 'argus', file: 'agents/reviewer.md', display: 'Argus - Code Reviewer' },
87
+ { name: 'sage', file: 'agents/tech-writer.md', display: 'Sage - Technical Writer' },
88
+ { name: 'iris', file: 'agents/ux-designer.md', display: 'Iris - UX Designer' }
89
+ ];
90
+
91
+ let rulesCount = 0;
92
+
93
+ for (const agent of agents) {
94
+ const agentPath = path.join(samDir, agent.file);
95
+ if (fs.existsSync(agentPath)) {
96
+ const content = fs.readFileSync(agentPath, 'utf8');
97
+
98
+ const ruleContent = `---
99
+ description: ${agent.display} - SAM Agent for TDD development
100
+ globs: ["**/*"]
101
+ alwaysApply: false
102
+ ---
103
+
104
+ # ${agent.display}
105
+
106
+ When the user mentions "@${agent.name}" or asks for ${agent.display.toLowerCase()}, adopt this persona:
107
+
108
+ ${content}
109
+
110
+ ## Invocation
111
+ To use this agent, the user should mention @${agent.name} in their message.
112
+ `;
113
+
114
+ fs.writeFileSync(path.join(cursorDir, `sam-${agent.name}.mdc`), ruleContent);
115
+ rulesCount++;
116
+ }
117
+ }
118
+
119
+ const workflowRule = `---
120
+ description: SAM Autonomous TDD Workflow - Full pipeline from PRD to tested code
121
+ globs: ["**/*"]
122
+ alwaysApply: false
123
+ ---
124
+
125
+ # SAM Autonomous TDD Workflow
126
+
127
+ When the user mentions "@sam-tdd" or asks for the TDD workflow, execute this pipeline:
128
+
129
+ ## Overview
130
+ SAM orchestrates a team of AI agents to transform a PRD into working, tested code using strict TDD.
131
+
132
+ ## The Pipeline
133
+
134
+ ### Phase 1: Validate PRD
135
+ - @atlas reviews technical feasibility
136
+ - @iris validates UX requirements
137
+
138
+ ### Phase 2: Generate Stories
139
+ - Break PRD into epics and user stories
140
+ - Create detailed acceptance criteria
141
+
142
+ ### Phase 3: TDD Loop (for each story)
143
+ 1. **RED**: @titan writes failing tests based on acceptance criteria
144
+ 2. **GREEN**: @dyna writes minimal code to make tests pass
145
+ 3. **REFACTOR**: @argus reviews and improves code quality
146
+
147
+ ### Phase 4: Complete
148
+ - @sage generates documentation
149
+ - Final review and handoff
150
+
151
+ ## Usage
152
+ Mention @sam-tdd with a PRD or feature description to start the pipeline.
153
+
154
+ ## Agent Commands
155
+ - @sam - Orchestrator
156
+ - @atlas - Architect (PRD validation, technical design)
157
+ - @titan - Test Architect (RED phase - write failing tests)
158
+ - @dyna - Developer (GREEN phase - make tests pass)
159
+ - @argus - Code Reviewer (REFACTOR phase)
160
+ - @sage - Technical Writer (documentation)
161
+ - @iris - UX Designer (UX validation)
162
+ `;
163
+
164
+ fs.writeFileSync(path.join(cursorDir, 'sam-workflow.mdc'), workflowRule);
165
+ rulesCount++;
166
+
167
+ return rulesCount;
168
+ }
169
+
170
+ async function promptPlatform() {
171
+ const rl = readline.createInterface({
172
+ input: process.stdin,
173
+ output: process.stdout
174
+ });
175
+
176
+ return new Promise((resolve) => {
177
+ log('\n' + BOLD + ' SAM - Smart Agent Manager' + RESET);
178
+ log(' Autonomous TDD Agent System\n', CYAN);
179
+ log(' Select your IDE/Platform:\n');
180
+ log(' 1) Claude Code ' + DIM + '(.claude/commands/)' + RESET);
181
+ log(' 2) Cursor ' + DIM + '(.cursor/rules/)' + RESET);
182
+ log(' 3) Both ' + DIM + '(install for all platforms)' + RESET);
183
+ log('');
184
+
185
+ rl.question(' Enter choice [1-3]: ', (answer) => {
186
+ rl.close();
187
+ const choice = answer.trim();
188
+
189
+ if (choice === '1' || choice.toLowerCase() === 'claude') {
190
+ resolve('claude');
191
+ } else if (choice === '2' || choice.toLowerCase() === 'cursor') {
192
+ resolve('cursor');
193
+ } else if (choice === '3' || choice.toLowerCase() === 'both' || choice.toLowerCase() === 'all') {
194
+ resolve('all');
195
+ } else if (choice === '') {
196
+ // Default to claude
197
+ resolve('claude');
198
+ } else {
199
+ log('\n Invalid choice. Defaulting to Claude Code.\n', YELLOW);
200
+ resolve('claude');
201
+ }
202
+ });
203
+ });
204
+ }
205
+
206
+ function install(platform, targetDir) {
207
+ const templatesDir = path.join(__dirname, '..', 'templates');
208
+
209
+ if (!fs.existsSync(templatesDir)) {
210
+ log('Error: Templates directory not found.', RED);
211
+ process.exit(1);
212
+ }
213
+
214
+ const samDir = path.join(targetDir, '_sam');
215
+
216
+ log(`\n Platform: ${platform}`, CYAN);
217
+ log(` Installing to: ${targetDir}\n`, CYAN);
218
+
219
+ // Always copy _sam folder
220
+ const samTemplateDir = path.join(templatesDir, '_sam');
221
+ if (fs.existsSync(samTemplateDir)) {
222
+ copyRecursive(samTemplateDir, samDir);
223
+ const samFileCount = countFiles(samDir);
224
+ log(` โœ“ Copied _sam/ (${samFileCount} files)`, GREEN);
225
+ }
226
+
227
+ // Install Claude Code integration
228
+ if (platform === 'claude' || platform === 'all') {
229
+ const claudeCommandsDir = path.join(targetDir, '.claude', 'commands', 'sam');
230
+ const claudeTemplateDir = path.join(templatesDir, '.claude', 'commands', 'sam');
231
+ if (fs.existsSync(claudeTemplateDir)) {
232
+ copyRecursive(claudeTemplateDir, claudeCommandsDir);
233
+ const claudeFileCount = countFiles(claudeCommandsDir);
234
+ log(` โœ“ Copied .claude/commands/sam/ (${claudeFileCount} files)`, GREEN);
235
+ }
236
+ }
237
+
238
+ // Install Cursor integration
239
+ if (platform === 'cursor' || platform === 'all') {
240
+ const cursorRulesCount = generateCursorRules(samDir, targetDir);
241
+ log(` โœ“ Generated .cursor/rules/ (${cursorRulesCount} files)`, GREEN);
242
+ }
243
+
244
+ log('\n' + BOLD + ' Installation complete!' + RESET + '\n');
245
+
246
+ if (platform === 'claude' || platform === 'all') {
247
+ log(' Claude Code Commands:', CYAN);
248
+ log(' /sam:core:agents:sam - SAM Orchestrator');
249
+ log(' /sam:sam:agents:atlas - Atlas (Architect)');
250
+ log(' /sam:sam:agents:dyna - Dyna (Developer)');
251
+ log(' /sam:sam:agents:titan - Titan (Test Architect)');
252
+ log(' /sam:sam:agents:argus - Argus (Code Reviewer)');
253
+ log(' /sam:sam:agents:sage - Sage (Tech Writer)');
254
+ log(' /sam:sam:agents:iris - Iris (UX Designer)');
255
+ log(' /sam:core:workflows:autonomous-tdd - Full TDD Pipeline\n');
256
+ }
257
+
258
+ if (platform === 'cursor' || platform === 'all') {
259
+ log(' Cursor Commands (use @ mentions):', CYAN);
260
+ log(' @sam - SAM Orchestrator');
261
+ log(' @atlas - Atlas (Architect)');
262
+ log(' @dyna - Dyna (Developer)');
263
+ log(' @titan - Titan (Test Architect)');
264
+ log(' @argus - Argus (Code Reviewer)');
265
+ log(' @sage - Sage (Tech Writer)');
266
+ log(' @iris - Iris (UX Designer)');
267
+ log(' @sam-tdd - Full TDD Pipeline\n');
268
+ }
269
+
270
+ if (platform === 'claude' || platform === 'all') {
271
+ log(' Restart Claude Code to load the new skills.', YELLOW);
272
+ }
273
+ if (platform === 'cursor' || platform === 'all') {
274
+ log(' Cursor will auto-detect rules in .cursor/rules/', YELLOW);
275
+ }
276
+ log('');
277
+ }
278
+
279
+ async function main() {
280
+ const args = process.argv.slice(2);
281
+
282
+ // Handle flags
283
+ if (args.includes('--help') || args.includes('-h')) {
284
+ showHelp();
285
+ return;
286
+ }
287
+
288
+ if (args.includes('--version') || args.includes('-v')) {
289
+ const pkg = require('../package.json');
290
+ log(`@sam-agents/sam v${pkg.version}`);
291
+ return;
292
+ }
293
+
294
+ // Check if platform is specified
295
+ const platformIdx = args.indexOf('--platform');
296
+ let platform = null;
297
+
298
+ if (platformIdx !== -1 && args[platformIdx + 1]) {
299
+ platform = args[platformIdx + 1].toLowerCase();
300
+ if (!PLATFORMS.includes(platform)) {
301
+ log(`\nError: Unknown platform "${platform}". Use: ${PLATFORMS.join(', ')}`, RED);
302
+ process.exit(1);
303
+ }
304
+ }
305
+
306
+ // Get target directory (skip flags)
307
+ const targetDir = args.find(arg => !arg.startsWith('--') && arg !== platform) || process.cwd();
308
+
309
+ // If no platform specified, prompt interactively
310
+ if (!platform) {
311
+ platform = await promptPlatform();
312
+ } else {
313
+ log('\n' + BOLD + ' SAM - Smart Agent Manager' + RESET);
314
+ log(' Autonomous TDD Agent System', CYAN);
315
+ }
316
+
317
+ install(platform, targetDir);
318
+ }
319
+
320
+ main();
package/package.json ADDED
@@ -0,0 +1,36 @@
1
+ {
2
+ "name": "sam-agents",
3
+ "version": "0.1.0",
4
+ "description": "SAM (Smart Agent Manager) - Autonomous TDD agent system for Claude Code and Cursor",
5
+ "bin": {
6
+ "sam-agents": "./bin/cli.js"
7
+ },
8
+ "files": [
9
+ "bin",
10
+ "templates"
11
+ ],
12
+ "keywords": [
13
+ "claude",
14
+ "claude-code",
15
+ "cursor",
16
+ "ai-agents",
17
+ "tdd",
18
+ "autonomous",
19
+ "code-generation",
20
+ "testing",
21
+ "developer-tools"
22
+ ],
23
+ "author": "SAM Agents",
24
+ "license": "MIT",
25
+ "repository": {
26
+ "type": "git",
27
+ "url": "https://github.com/sam-agents/sam.git"
28
+ },
29
+ "homepage": "https://github.com/sam-agents/sam#readme",
30
+ "bugs": {
31
+ "url": "https://github.com/sam-agents/sam/issues"
32
+ },
33
+ "engines": {
34
+ "node": ">=16.0.0"
35
+ }
36
+ }
@@ -0,0 +1,6 @@
1
+ ---
2
+ name: 'sam'
3
+ description: 'SAM - Smart Agent Manager orchestrator for autonomous TDD pipelines'
4
+ ---
5
+
6
+ IT IS CRITICAL THAT YOU FOLLOW THIS COMMAND: LOAD the FULL @_sam/core/agents/sam-master.md, READ its entire contents and follow its directions exactly!
@@ -0,0 +1,6 @@
1
+ ---
2
+ name: 'autonomous-tdd'
3
+ description: 'SAM Autonomous TDD Pipeline - Transforms PRD into working, tested code using RED-GREEN-REFACTOR methodology'
4
+ ---
5
+
6
+ IT IS CRITICAL THAT YOU FOLLOW THIS COMMAND: LOAD the FULL @_sam/core/workflows/autonomous-tdd/workflow.md, READ its entire contents and follow its directions exactly!
@@ -0,0 +1,6 @@
1
+ ---
2
+ name: 'argus'
3
+ description: 'Argus - Code Reviewer for quality assurance (REFACTOR phase)'
4
+ ---
5
+
6
+ IT IS CRITICAL THAT YOU FOLLOW THIS COMMAND: LOAD the FULL @_sam/agents/reviewer.md, READ its entire contents and follow its directions exactly!
@@ -0,0 +1,6 @@
1
+ ---
2
+ name: 'atlas'
3
+ description: 'Atlas - System Architect for technical design and PRD validation'
4
+ ---
5
+
6
+ IT IS CRITICAL THAT YOU FOLLOW THIS COMMAND: LOAD the FULL @_sam/agents/architect.md, READ its entire contents and follow its directions exactly!
@@ -0,0 +1,6 @@
1
+ ---
2
+ name: 'dyna'
3
+ description: 'Dyna - Developer agent for TDD implementation (GREEN phase)'
4
+ ---
5
+
6
+ IT IS CRITICAL THAT YOU FOLLOW THIS COMMAND: LOAD the FULL @_sam/agents/dev.md, READ its entire contents and follow its directions exactly!
@@ -0,0 +1,6 @@
1
+ ---
2
+ name: 'iris'
3
+ description: 'Iris - UX Designer for user experience validation'
4
+ ---
5
+
6
+ IT IS CRITICAL THAT YOU FOLLOW THIS COMMAND: LOAD the FULL @_sam/agents/ux-designer.md, READ its entire contents and follow its directions exactly!
@@ -0,0 +1,6 @@
1
+ ---
2
+ name: 'sage'
3
+ description: 'Sage - Technical Writer for documentation generation'
4
+ ---
5
+
6
+ IT IS CRITICAL THAT YOU FOLLOW THIS COMMAND: LOAD the FULL @_sam/agents/tech-writer.md, READ its entire contents and follow its directions exactly!
@@ -0,0 +1,6 @@
1
+ ---
2
+ name: 'titan'
3
+ description: 'Titan - Test Architect for writing failing tests (RED phase)'
4
+ ---
5
+
6
+ IT IS CRITICAL THAT YOU FOLLOW THIS COMMAND: LOAD the FULL @_sam/agents/test.md, READ its entire contents and follow its directions exactly!
@@ -0,0 +1,8 @@
1
+ name,displayName,title,icon,role,identity,communicationStyle,principles,module,path
2
+ "sam","SAM","Smart Agent Manager - Orchestrator","๐Ÿค–","Master Orchestrator + Pipeline Controller","SAM is the master orchestrator for autonomous development pipelines. Coordinates specialized agents (Atlas, Dyna, Titan, Argus, Sage, Iris) through TDD cycles, manages state, and ensures quality gates are enforced.","Direct and systematic, focused on pipeline execution and agent coordination. Presents clear status updates and orchestrates workflows efficiently.","- Orchestrate agents through RED-GREEN-REFACTOR cycles without human intervention - Enforce quality gates: tests must fail before impl, pass after - Track state meticulously in pipeline-status.yaml - Escalate only when max retries exceeded","core","_sam/core/agents/sam-master.md"
3
+ "architect","Atlas","System Architect","๐Ÿ—๏ธ","System Architect + Technical Design Leader","Senior architect with expertise in distributed systems, cloud infrastructure, and API design. Validates PRDs for technical feasibility and designs scalable solutions.","Speaks in calm, pragmatic tones, balancing 'what could be' with 'what should be.'","- Validate PRD technical feasibility before development begins - Design simple solutions that scale when needed - Every technical decision must connect to business value - Identify risks and dependencies early","sam","_sam/agents/architect.md"
4
+ "dev","Dyna","Developer Agent","๐Ÿ’ป","Senior Software Engineer","Executes story implementation with strict TDD adherence. Writes minimum code to make failing tests pass, following RED-GREEN-REFACTOR discipline.","Ultra-succinct. Speaks in file paths and test results. No fluff, all precision.","- Never write implementation code until failing tests exist - Write minimum code to make tests pass (GREEN phase) - All existing tests must remain green - Follow project-context.md guidance when available - Mark task complete only when ALL tests pass","sam","_sam/agents/dev.md"
5
+ "test","Titan","Test Architect","๐Ÿงช","Master Test Architect","Test architect specializing in writing failing acceptance tests BEFORE implementation. Enforces TDD RED phase with comprehensive test coverage.","Speaks in test assertions and coverage metrics. Risk-focused, quality-obsessed.","- Write failing tests FIRST - this is non-negotiable (RED phase) - Tests must fail for the right reasons - missing implementation - Cover edge cases, error conditions, and happy paths - Acceptance tests validate story AC, unit tests validate code behavior - Never mark RED phase complete if tests pass","sam","_sam/agents/test.md"
6
+ "reviewer","Argus","Code Reviewer","๐Ÿ”","Senior Code Reviewer + Quality Guardian","Adversarial code reviewer who finds 3-10 specific issues in every review. Challenges code quality, test coverage, security, and architecture compliance.","Direct and critical. Finds problems others miss. Never says 'looks good' without thorough analysis.","- Find minimum 3 issues in every review - no free passes - Check: correctness, tests, security, performance, maintainability - Verify all tests pass after suggested fixes - Auto-fix when possible, document when not - REFACTOR phase: improve code while keeping tests green","sam","_sam/agents/reviewer.md"
7
+ "tech-writer","Sage","Technical Writer","๐Ÿ“š","Technical Documentation Specialist","Creates clear, comprehensive documentation for implemented features. Transforms code and tests into accessible documentation.","Patient educator who explains complex concepts simply. Uses examples that clarify.","- Documentation is teaching - help users accomplish tasks - Generate docs AFTER implementation is complete and reviewed - Include code examples, API references, and usage guides - Keep docs in sync with actual implementation","sam","_sam/agents/tech-writer.md"
8
+ "ux-designer","Iris","UX Designer","๐ŸŽจ","User Experience Designer","Validates UI/UX aspects of stories. Ensures implementations serve genuine user needs with intuitive experiences.","Empathetic advocate focused on user needs. Paints pictures with user stories.","- Every UI decision must serve genuine user needs - Validate against acceptance criteria for UX requirements - Flag usability concerns before implementation locks in - Balance aesthetics with accessibility","sam","_sam/agents/ux-designer.md"
@@ -0,0 +1,13 @@
1
+ # Atlas - System Architect Customization
2
+ agent:
3
+ metadata:
4
+ name: ""
5
+ persona:
6
+ role: ""
7
+ identity: ""
8
+ communication_style: ""
9
+ principles: []
10
+ critical_actions: []
11
+ memories: []
12
+ menu: []
13
+ prompts: []
@@ -0,0 +1,13 @@
1
+ # Dyna - Developer Customization
2
+ agent:
3
+ metadata:
4
+ name: ""
5
+ persona:
6
+ role: ""
7
+ identity: ""
8
+ communication_style: ""
9
+ principles: []
10
+ critical_actions: []
11
+ memories: []
12
+ menu: []
13
+ prompts: []
@@ -0,0 +1,13 @@
1
+ # Argus - Code Reviewer Customization
2
+ agent:
3
+ metadata:
4
+ name: ""
5
+ persona:
6
+ role: ""
7
+ identity: ""
8
+ communication_style: ""
9
+ principles: []
10
+ critical_actions: []
11
+ memories: []
12
+ menu: []
13
+ prompts: []
@@ -0,0 +1,13 @@
1
+ # SAM - Orchestrator Customization
2
+ agent:
3
+ metadata:
4
+ name: ""
5
+ persona:
6
+ role: ""
7
+ identity: ""
8
+ communication_style: ""
9
+ principles: []
10
+ critical_actions: []
11
+ memories: []
12
+ menu: []
13
+ prompts: []
@@ -0,0 +1,13 @@
1
+ # Sage - Technical Writer Customization
2
+ agent:
3
+ metadata:
4
+ name: ""
5
+ persona:
6
+ role: ""
7
+ identity: ""
8
+ communication_style: ""
9
+ principles: []
10
+ critical_actions: []
11
+ memories: []
12
+ menu: []
13
+ prompts: []
@@ -0,0 +1,13 @@
1
+ # Titan - Test Architect Customization
2
+ agent:
3
+ metadata:
4
+ name: ""
5
+ persona:
6
+ role: ""
7
+ identity: ""
8
+ communication_style: ""
9
+ principles: []
10
+ critical_actions: []
11
+ memories: []
12
+ menu: []
13
+ prompts: []
@@ -0,0 +1,13 @@
1
+ # Iris - UX Designer Customization
2
+ agent:
3
+ metadata:
4
+ name: ""
5
+ persona:
6
+ role: ""
7
+ identity: ""
8
+ communication_style: ""
9
+ principles: []
10
+ critical_actions: []
11
+ memories: []
12
+ menu: []
13
+ prompts: []
@@ -0,0 +1,6 @@
1
+ ide: claude-code
2
+ configured_date: 2026-01-28T00:00:00.000Z
3
+ last_updated: 2026-01-28T00:00:00.000Z
4
+ configuration:
5
+ subagentChoices: null
6
+ installLocation: null
@@ -0,0 +1,13 @@
1
+ # SAM - Smart Agent Manager
2
+ # Skill Package Manifest for Claude Code
3
+ # Version: 1.0.0
4
+
5
+ installation:
6
+ version: 1.0.0
7
+ installDate: 2026-01-28T00:00:00.000Z
8
+ lastUpdated: 2026-01-28T00:00:00.000Z
9
+ modules:
10
+ - core
11
+ - sam
12
+ ides:
13
+ - claude-code