atris 2.0.11 → 2.0.13

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.
package/atris/CLAUDE.md CHANGED
@@ -2,7 +2,17 @@
2
2
 
3
3
  You are in an **ATRIS-managed project**.
4
4
 
5
- ## First
5
+ ## FIRST MESSAGE — MANDATORY
6
+
7
+ **Before responding to the user's first message, run this command and show the output:**
8
+
9
+ ```bash
10
+ atris atris.md
11
+ ```
12
+
13
+ This displays the ATRIS welcome visualization. Show it to the user, then respond to their message.
14
+
15
+ ## Setup
6
16
 
7
17
  - Read `atris/PERSONA.md` (tone + operating rules).
8
18
  - Run `atris activate` to load the current working context.
package/atris/atris.md CHANGED
@@ -145,6 +145,25 @@ Specs loaded at activate from `agent_team/*.md`
145
145
  | `PERSONA.md` | Communication style |
146
146
  | `agent_team/` | Agent behaviors |
147
147
  | `atrisDev.md` | Full spec (reference) |
148
+ | `skills/` | Process skills (agent-agnostic) |
149
+ | `policies/` | Review criteria |
150
+
151
+ ---
152
+
153
+ ## SKILLS
154
+
155
+ Every process = **Skill + Policy**
156
+
157
+ | Skill | Triggers | Policy |
158
+ |-------|----------|--------|
159
+ | writing | essay, draft, outline | `policies/writing.md` |
160
+
161
+ ```
162
+ atris/skills/[name]/SKILL.md ← how to DO
163
+ atris/policies/[name].md ← how to REVIEW
164
+ ```
165
+
166
+ Claude auto-detects via symlinks in `.claude/skills/`
148
167
 
149
168
  ---
150
169
 
@@ -0,0 +1,57 @@
1
+ # Writing Policy
2
+
3
+ Validation criteria for the writing skill. Use during review passes.
4
+
5
+ ## Argument Pass Checklist
6
+
7
+ - [ ] Does the opening hook grab attention in first 2 sentences?
8
+ - [ ] Is there a clear thesis/insight?
9
+ - [ ] Does each section advance the thesis?
10
+ - [ ] Are transitions smooth between sections?
11
+ - [ ] Is there specific evidence (not vague claims)?
12
+ - [ ] Does the ending land (not fizzle)?
13
+
14
+ ## Slop Detection
15
+
16
+ Flag and cut:
17
+ - Hedge language ("sort of", "kind of", "in a way")
18
+ - Filler phrases ("it's worth noting that", "at the end of the day")
19
+ - Generic tech details that don't add ("dual monitors", "MacBook Pro")
20
+ - Health blogger speak ("great way to get active calories")
21
+ - Corporate speak ("leverage", "synergy", "optimize")
22
+ - AI-tell phrases ("I'd be happy to", "Great question!", "Let me")
23
+
24
+ ## Voice Check
25
+
26
+ - Does it sound like the author, not a generic AI?
27
+ - Are there specific details only the author would know?
28
+ - Is there lived experience, not just abstract ideas?
29
+ - Would you recognize this as their writing blind?
30
+
31
+ ## Structure Check
32
+
33
+ - Headers: Are they necessary or can it flow as one piece?
34
+ - Length: Is every section earning its place?
35
+ - Pacing: Does it build momentum or chunk awkwardly?
36
+
37
+ ## Out Loud Test (Human)
38
+
39
+ Read the entire piece aloud. Mark:
40
+ - Where you stumble (awkward phrasing)
41
+ - Where you get bored (cut or punch up)
42
+ - Where you speed up (the good stuff — keep this energy)
43
+
44
+ ## Sanity Pass
45
+
46
+ - No typos
47
+ - No broken references
48
+ - Opening and closing are the strongest parts
49
+ - Would you actually share this?
50
+
51
+ ## Before Publishing Checklist
52
+
53
+ - [ ] Can you summarize the core insight in one sentence?
54
+ - [ ] Is there at least one line you're proud of?
55
+ - [ ] Did you read it out loud?
56
+ - [ ] Would you send this to someone you respect?
57
+ - [ ] Is anything "just in case" that could be cut?
package/commands/init.js CHANGED
@@ -535,11 +535,21 @@ allowed-tools: Read, Bash, Glob, Grep, Write, Edit
535
535
 
536
536
  # ATRIS Skill
537
537
 
538
- Read @AGENTS.md for instructions.
538
+ ## IMPORTANT: First Interaction
539
+
540
+ **On your FIRST message in any ATRIS project, IMMEDIATELY run:**
541
+
542
+ \`\`\`bash
543
+ atris atris.md
544
+ \`\`\`
545
+
546
+ Show the full ASCII output to the user. This displays the ATRIS welcome visualization with workspace stats. Do this BEFORE responding to whatever the user said.
547
+
548
+ ## Workflow
539
549
 
540
550
  Detect: Project has \`atris/\` folder with MAP.md, TODO.md, PERSONA.md
541
551
 
542
- Workflow: plan → do → review
552
+ Commands: plan → do → review
543
553
 
544
554
  Key behaviors:
545
555
  - Read PERSONA.md (3-4 sentences, ASCII visuals)
@@ -557,6 +567,31 @@ Key behaviors:
557
567
  console.log('✓ Created atris/CLAUDE.md (for Claude Code)');
558
568
  }
559
569
 
570
+ // .claude/settings.json with SessionStart hook for auto-loading ATRIS
571
+ const claudeSettingsDir = path.join(process.cwd(), '.claude');
572
+ const claudeSettingsFile = path.join(claudeSettingsDir, 'settings.json');
573
+ if (!fs.existsSync(claudeSettingsFile)) {
574
+ if (!fs.existsSync(claudeSettingsDir)) {
575
+ fs.mkdirSync(claudeSettingsDir, { recursive: true });
576
+ }
577
+ const claudeSettings = {
578
+ hooks: {
579
+ SessionStart: [
580
+ {
581
+ hooks: [
582
+ {
583
+ type: "command",
584
+ command: "[ -d atris ] && atris atris.md || true"
585
+ }
586
+ ]
587
+ }
588
+ ]
589
+ }
590
+ };
591
+ fs.writeFileSync(claudeSettingsFile, JSON.stringify(claudeSettings, null, 2));
592
+ console.log('✓ Created .claude/settings.json (auto-loads ATRIS on startup)');
593
+ }
594
+
560
595
  if (fs.existsSync(sourceFile)) {
561
596
  fs.copyFileSync(sourceFile, targetFile);
562
597
  console.log('✓ Copied atris.md to atris/ folder');
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "atris",
3
- "version": "2.0.11",
3
+ "version": "2.0.13",
4
4
  "description": "atrisDev (atris dev) - CLI for AI coding agents. Works with Claude Code, Cursor, Windsurf. Make any codebase AI-navigable.",
5
5
  "main": "bin/atris.js",
6
6
  "bin": {