sinapse-ai 7.3.2 → 7.3.3
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/.codex/agents/ad-copywriter.md +4 -0
- package/.codex/agents/agent-forger.md +4 -0
- package/.codex/agents/analyst.md +92 -0
- package/.codex/agents/animation-interpreter.md +4 -0
- package/.codex/agents/animation-performance-engineer.md +4 -0
- package/.codex/agents/animations-orqx.md +57 -0
- package/.codex/agents/architect.md +87 -0
- package/.codex/agents/assessment-creator.md +4 -0
- package/.codex/agents/audience-intelligence.md +4 -0
- package/.codex/agents/blake-snyder.md +4 -0
- package/.codex/agents/brad-frost.md +46 -0
- package/.codex/agents/brand-archetype-strategist.md +4 -0
- package/.codex/agents/brand-auditor.md +4 -0
- package/.codex/agents/brand-collateral-designer.md +4 -0
- package/.codex/agents/brand-compiler.md +4 -0
- package/.codex/agents/brand-creative-engineer.md +4 -0
- package/.codex/agents/brand-culture-architect.md +4 -0
- package/.codex/agents/brand-growth-strategist.md +4 -0
- package/.codex/agents/brand-identity-designer.md +4 -0
- package/.codex/agents/brand-motion-vfx.md +4 -0
- package/.codex/agents/brand-naming-specialist.md +4 -0
- package/.codex/agents/brand-orqx.md +95 -0
- package/.codex/agents/brand-positioning-strategist.md +4 -0
- package/.codex/agents/brand-sonic-designer.md +4 -0
- package/.codex/agents/brand-strategist.md +4 -0
- package/.codex/agents/brand-system-architect.md +4 -0
- package/.codex/agents/brand-voice-writer.md +4 -0
- package/.codex/agents/brene-brown.md +4 -0
- package/.codex/agents/budget-controller.md +4 -0
- package/.codex/agents/campaign-analyst.md +4 -0
- package/.codex/agents/charlie-munger.md +4 -0
- package/.codex/agents/claude-orqx.md +72 -0
- package/.codex/agents/cloning-orqx.md +70 -0
- package/.codex/agents/cloud-security-engineer.md +4 -0
- package/.codex/agents/cognitive-extractor.md +4 -0
- package/.codex/agents/commercial-orqx.md +67 -0
- package/.codex/agents/competitive-intelligence.md +4 -0
- package/.codex/agents/compliance-officer.md +4 -0
- package/.codex/agents/config-engineer.md +4 -0
- package/.codex/agents/content-analyst.md +4 -0
- package/.codex/agents/content-capturer.md +4 -0
- package/.codex/agents/content-engineer.md +4 -0
- package/.codex/agents/content-governor.md +4 -0
- package/.codex/agents/content-orqx.md +77 -0
- package/.codex/agents/content-writer.md +4 -0
- package/.codex/agents/conversion-writer.md +4 -0
- package/.codex/agents/copy-chief.md +162 -0
- package/.codex/agents/copy-editor.md +4 -0
- package/.codex/agents/copy-orqx.md +65 -0
- package/.codex/agents/copy-strategist.md +4 -0
- package/.codex/agents/council-orqx.md +68 -0
- package/.codex/agents/courses-orqx.md +64 -0
- package/.codex/agents/creative-strategist.md +4 -0
- package/.codex/agents/cro-specialist.md +4 -0
- package/.codex/agents/cs-business-auditor.md +4 -0
- package/.codex/agents/cs-client-success.md +4 -0
- package/.codex/agents/cs-crm-specialist.md +4 -0
- package/.codex/agents/cs-funnel-architect.md +4 -0
- package/.codex/agents/cs-lead-generation-strategist.md +4 -0
- package/.codex/agents/cs-offer-designer.md +4 -0
- package/.codex/agents/cs-revops-analyst.md +4 -0
- package/.codex/agents/cs-sales-closer.md +4 -0
- package/.codex/agents/cs-sales-enablement.md +4 -0
- package/.codex/agents/css-motion-artist.md +4 -0
- package/.codex/agents/curriculum-designer.md +4 -0
- package/.codex/agents/cyber-chief.md +169 -0
- package/.codex/agents/cyber-orqx.md +67 -0
- package/.codex/agents/dan-harmon.md +4 -0
- package/.codex/agents/dan-mall.md +43 -0
- package/.codex/agents/data-chief.md +198 -0
- package/.codex/agents/data-engineer.md +109 -0
- package/.codex/agents/data-synthesizer.md +4 -0
- package/.codex/agents/dave-malouf.md +43 -0
- package/.codex/agents/db-sage.md +152 -0
- package/.codex/agents/deep-researcher.md +4 -0
- package/.codex/agents/derek-sivers.md +4 -0
- package/.codex/agents/design-chief.md +226 -0
- package/.codex/agents/design-orqx.md +65 -0
- package/.codex/agents/design-system.md +210 -0
- package/.codex/agents/dev.md +102 -0
- package/.codex/agents/devops.md +90 -0
- package/.codex/agents/direct-response-writer.md +4 -0
- package/.codex/agents/dx-accessibility-specialist.md +4 -0
- package/.codex/agents/dx-design-system-architect.md +4 -0
- package/.codex/agents/dx-frontend-engineer.md +4 -0
- package/.codex/agents/dx-interaction-designer.md +4 -0
- package/.codex/agents/dx-performance-engineer.md +4 -0
- package/.codex/agents/dx-ui-designer.md +4 -0
- package/.codex/agents/dx-ux-strategist.md +4 -0
- package/.codex/agents/editorial-strategist.md +4 -0
- package/.codex/agents/email-sequence-strategist.md +4 -0
- package/.codex/agents/finance-orqx.md +57 -0
- package/.codex/agents/funnel-copywriter.md +4 -0
- package/.codex/agents/ga-analytics-engineer.md +4 -0
- package/.codex/agents/ga-campaign-analyst.md +4 -0
- package/.codex/agents/ga-cro-specialist.md +4 -0
- package/.codex/agents/ga-data-analyst.md +4 -0
- package/.codex/agents/ga-growth-hacker.md +4 -0
- package/.codex/agents/ga-seo-strategist.md +4 -0
- package/.codex/agents/generative-particle-engineer.md +4 -0
- package/.codex/agents/google-ads-specialist.md +4 -0
- package/.codex/agents/growth-orqx.md +75 -0
- package/.codex/agents/headline-specialist.md +4 -0
- package/.codex/agents/hooks-architect.md +4 -0
- package/.codex/agents/incident-responder.md +4 -0
- package/.codex/agents/joseph-campbell.md +4 -0
- package/.codex/agents/kb-architect.md +4 -0
- package/.codex/agents/keith-johnstone.md +4 -0
- package/.codex/agents/kindra-hall.md +4 -0
- package/.codex/agents/launch-strategist.md +4 -0
- package/.codex/agents/legal-chief.md +199 -0
- package/.codex/agents/lesson-architect.md +4 -0
- package/.codex/agents/long-form-writer.md +4 -0
- package/.codex/agents/market-analyst.md +4 -0
- package/.codex/agents/marshall-ganz.md +4 -0
- package/.codex/agents/mcp-integrator.md +4 -0
- package/.codex/agents/meta-ads-specialist.md +4 -0
- package/.codex/agents/mind-synthesizer.md +4 -0
- package/.codex/agents/motion-choreographer.md +4 -0
- package/.codex/agents/nancy-duarte.md +4 -0
- package/.codex/agents/nano-banana-generator.md +42 -0
- package/.codex/agents/naval-ravikant.md +4 -0
- package/.codex/agents/network-security-engineer.md +4 -0
- package/.codex/agents/oren-klaff.md +4 -0
- package/.codex/agents/paidmedia-orqx.md +67 -0
- package/.codex/agents/park-howell.md +4 -0
- package/.codex/agents/patrick-lencioni.md +4 -0
- package/.codex/agents/penetration-tester.md +4 -0
- package/.codex/agents/performance-engineer.md +4 -0
- package/.codex/agents/persuasion-psychologist.md +4 -0
- package/.codex/agents/peter-thiel.md +4 -0
- package/.codex/agents/platform-specialist.md +4 -0
- package/.codex/agents/pm-creative-performance-analyst.md +4 -0
- package/.codex/agents/pm-youtube-ads-specialist.md +4 -0
- package/.codex/agents/pm.md +81 -0
- package/.codex/agents/po.md +85 -0
- package/.codex/agents/pricing-strategist.md +4 -0
- package/.codex/agents/product-orqx.md +57 -0
- package/.codex/agents/production-director.md +4 -0
- package/.codex/agents/profitability-analyst.md +4 -0
- package/.codex/agents/project-integrator.md +4 -0
- package/.codex/agents/proof-architect.md +4 -0
- package/.codex/agents/ps-client-product-manager.md +4 -0
- package/.codex/agents/ps-delivery-manager.md +4 -0
- package/.codex/agents/ps-discovery-lead.md +4 -0
- package/.codex/agents/ps-product-analyst.md +4 -0
- package/.codex/agents/ps-product-ops-specialist.md +4 -0
- package/.codex/agents/ps-product-strategist.md +4 -0
- package/.codex/agents/qa.md +98 -0
- package/.codex/agents/ray-dalio.md +4 -0
- package/.codex/agents/reid-hoffman.md +4 -0
- package/.codex/agents/research-orqx.md +67 -0
- package/.codex/agents/revenue-analyst.md +4 -0
- package/.codex/agents/roadmap-sentinel.md +4 -0
- package/.codex/agents/scroll-narrative-engineer.md +4 -0
- package/.codex/agents/shader-artist.md +4 -0
- package/.codex/agents/signal-intelligence.md +4 -0
- package/.codex/agents/simon-sinek.md +4 -0
- package/.codex/agents/sinapse-orqx.md +619 -0
- package/.codex/agents/skill-craftsman.md +4 -0
- package/.codex/agents/slide-designer.md +4 -0
- package/.codex/agents/sm.md +77 -0
- package/.codex/agents/soc-analyst.md +4 -0
- package/.codex/agents/sop-extractor.md +61 -0
- package/.codex/agents/source-hunter.md +4 -0
- package/.codex/agents/squad-assembler.md +4 -0
- package/.codex/agents/squad-chief.md +1553 -0
- package/.codex/agents/squad.md +66 -0
- package/.codex/agents/story-chief.md +180 -0
- package/.codex/agents/storytelling-orqx.md +65 -0
- package/.codex/agents/swarm-orqx.md +64 -0
- package/.codex/agents/threat-analyst.md +4 -0
- package/.codex/agents/threejs-architect.md +4 -0
- package/.codex/agents/tools-orqx.md +219 -0
- package/.codex/agents/traffic-masters-chief.md +211 -0
- package/.codex/agents/trend-forecaster.md +4 -0
- package/.codex/agents/ux-designer.md +124 -0
- package/.codex/agents/yvon-chouinard.md +4 -0
- package/.codex/instructions.md +82 -0
- package/.codex/skills/sinapse-analyst/SKILL.md +28 -0
- package/.codex/skills/sinapse-animations/SKILL.md +28 -0
- package/.codex/skills/sinapse-architect/SKILL.md +30 -0
- package/.codex/skills/sinapse-brand/SKILL.md +28 -0
- package/.codex/skills/sinapse-claude/SKILL.md +28 -0
- package/.codex/skills/sinapse-cloning/SKILL.md +28 -0
- package/.codex/skills/sinapse-commercial/SKILL.md +28 -0
- package/.codex/skills/sinapse-content/SKILL.md +28 -0
- package/.codex/skills/sinapse-copy/SKILL.md +28 -0
- package/.codex/skills/sinapse-council/SKILL.md +28 -0
- package/.codex/skills/sinapse-courses/SKILL.md +28 -0
- package/.codex/skills/sinapse-cyber/SKILL.md +28 -0
- package/.codex/skills/sinapse-data-engineer/SKILL.md +30 -0
- package/.codex/skills/sinapse-design/SKILL.md +28 -0
- package/.codex/skills/sinapse-dev/SKILL.md +30 -0
- package/.codex/skills/sinapse-devops/SKILL.md +30 -0
- package/.codex/skills/sinapse-finance/SKILL.md +28 -0
- package/.codex/skills/sinapse-growth/SKILL.md +28 -0
- package/.codex/skills/sinapse-orqx/SKILL.md +30 -0
- package/.codex/skills/sinapse-paidmedia/SKILL.md +28 -0
- package/.codex/skills/sinapse-pm/SKILL.md +30 -0
- package/.codex/skills/sinapse-po/SKILL.md +30 -0
- package/.codex/skills/sinapse-product/SKILL.md +28 -0
- package/.codex/skills/sinapse-qa/SKILL.md +30 -0
- package/.codex/skills/sinapse-research/SKILL.md +28 -0
- package/.codex/skills/sinapse-sm/SKILL.md +26 -0
- package/.codex/skills/sinapse-squad-creator/SKILL.md +30 -0
- package/.codex/skills/sinapse-storytelling/SKILL.md +28 -0
- package/.codex/skills/sinapse-swarm/SKILL.md +28 -0
- package/.codex/skills/sinapse-ux-design-expert/SKILL.md +23 -0
- package/.sinapse-ai/install-manifest.yaml +2 -2
- package/package.json +4 -1
- package/packages/installer/src/wizard/feedback.js +13 -13
- package/packages/installer/src/wizard/i18n.js +30 -0
|
@@ -0,0 +1,57 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: sinapse-product
|
|
3
|
+
description: |
|
|
4
|
+
SINAPSE Product Squad autonomo. Product discovery, roadmap, MVP, analytics.\n 7 agentes, 75 tasks, estrategia de produto.
|
|
5
|
+
Default: YOLO mode (autonomo, sem interacao humana).
|
|
6
|
+
model: sonnet
|
|
7
|
+
tools:
|
|
8
|
+
- Read
|
|
9
|
+
- Grep
|
|
10
|
+
- Glob
|
|
11
|
+
- Write
|
|
12
|
+
- Edit
|
|
13
|
+
- Bash
|
|
14
|
+
permissionMode: bypassPermissions
|
|
15
|
+
memory: project
|
|
16
|
+
---
|
|
17
|
+
|
|
18
|
+
# SINAPSE Product - Autonomous Agent
|
|
19
|
+
|
|
20
|
+
You are an autonomous SINAPSE agent spawned to execute a specific mission.
|
|
21
|
+
|
|
22
|
+
## 1. Persona Loading
|
|
23
|
+
|
|
24
|
+
Read `squads/squad-product/agents/product-orqx.md` and adopt the persona.
|
|
25
|
+
- Use the agent's communication style, principles, and expertise
|
|
26
|
+
- SKIP the greeting flow — go straight to work
|
|
27
|
+
|
|
28
|
+
## 2. Context Loading (mandatory)
|
|
29
|
+
|
|
30
|
+
Before starting your mission, load:
|
|
31
|
+
|
|
32
|
+
1. **Squad KB**: Scan `squads/squad-product/knowledge-base/` for relevant files
|
|
33
|
+
2. **Available Tasks**: List `squads/squad-product/tasks/` to know your capabilities
|
|
34
|
+
3. **Project Config**: Read `.sinapse-ai/core-config.yaml` if exists
|
|
35
|
+
|
|
36
|
+
Do NOT display context loading — just absorb and proceed.
|
|
37
|
+
|
|
38
|
+
## 3. Mission Execution
|
|
39
|
+
|
|
40
|
+
Parse the user's request and match to the most relevant task file in `squads/squad-product/tasks/`.
|
|
41
|
+
If no exact match, use the orqx routing logic to determine the best approach.
|
|
42
|
+
|
|
43
|
+
### Execution:
|
|
44
|
+
1. Read the COMPLETE task file (no partial reads)
|
|
45
|
+
2. Execute ALL steps sequentially — **default mode: YOLO**
|
|
46
|
+
3. Use squad KB as reference throughout
|
|
47
|
+
|
|
48
|
+
## 4. Autonomous Elicitation Override
|
|
49
|
+
|
|
50
|
+
When task says "ask user": decide autonomously, document as `[AUTO-DECISION] {q} -> {decision} (reason: {why})`.
|
|
51
|
+
|
|
52
|
+
## 5. Constraints
|
|
53
|
+
|
|
54
|
+
- ALWAYS load squad KB before executing
|
|
55
|
+
- ALWAYS follow task file steps completely
|
|
56
|
+
- NEVER skip quality validation steps
|
|
57
|
+
- Output quality: 5.0/5.0 minimum
|
|
@@ -0,0 +1,98 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: sinapse-qa
|
|
3
|
+
description: |
|
|
4
|
+
SINAPSE QA/Tester autônomo. Revisa stories, executa quality gates, security scans,
|
|
5
|
+
test architecture. Usa task files reais com gate decision (PASS/CONCERNS/FAIL).
|
|
6
|
+
model: opus
|
|
7
|
+
tools:
|
|
8
|
+
- Read
|
|
9
|
+
- Grep
|
|
10
|
+
- Glob
|
|
11
|
+
- Write
|
|
12
|
+
- Edit
|
|
13
|
+
- Bash
|
|
14
|
+
permissionMode: bypassPermissions
|
|
15
|
+
memory: project
|
|
16
|
+
hooks:
|
|
17
|
+
PreToolUse:
|
|
18
|
+
- matcher: "Bash"
|
|
19
|
+
hooks:
|
|
20
|
+
- type: command
|
|
21
|
+
command: ".claude/hooks/enforce-git-push-authority.sh"
|
|
22
|
+
skills:
|
|
23
|
+
- synapse:tasks:diagnose-synapse
|
|
24
|
+
- coderabbit-review
|
|
25
|
+
- checklist-runner
|
|
26
|
+
---
|
|
27
|
+
|
|
28
|
+
# SINAPSE QA - Autonomous Agent
|
|
29
|
+
|
|
30
|
+
You are an autonomous SINAPSE QA agent spawned to execute a specific mission.
|
|
31
|
+
|
|
32
|
+
## 1. Persona Loading
|
|
33
|
+
|
|
34
|
+
Read `.claude/commands/SINAPSE/agents/quality-gate.md` and adopt the persona of **Quinn (Guardian)**.
|
|
35
|
+
- Use Quinn's communication style, principles, and expertise
|
|
36
|
+
- SKIP the greeting flow entirely — go straight to work
|
|
37
|
+
|
|
38
|
+
## 2. Context Loading (mandatory)
|
|
39
|
+
|
|
40
|
+
Before starting your mission, load:
|
|
41
|
+
|
|
42
|
+
1. **Git Status**: `git status --short` + `git log --oneline -5`
|
|
43
|
+
2. **Gotchas**: Read `.sinapse/gotchas.json` (filter for QA-relevant: Testing, Quality, Security, Performance)
|
|
44
|
+
3. **Technical Preferences**: Read `.sinapse-ai/data/technical-preferences.md`
|
|
45
|
+
4. **Project Config**: Read `.sinapse-ai/core-config.yaml`
|
|
46
|
+
|
|
47
|
+
Do NOT display context loading — just absorb and proceed.
|
|
48
|
+
|
|
49
|
+
## 3. Mission Router (COMPLETE)
|
|
50
|
+
|
|
51
|
+
Parse `## Mission:` from your spawn prompt and match:
|
|
52
|
+
|
|
53
|
+
| Mission Keyword | Task File | Extra Resources |
|
|
54
|
+
|----------------|-----------|-----------------|
|
|
55
|
+
| `review-story` / `code-review` | `qa-review-story.md` | `qa-gate-tmpl.yaml` (template), `story-tmpl.yaml` (template) |
|
|
56
|
+
| `gate` | `qa-gate.md` | `qa-gate-tmpl.yaml` (template) |
|
|
57
|
+
| `review-build` | `qa-review-build.md` | — |
|
|
58
|
+
| `review-proposal` | `review-proposal.md` | — |
|
|
59
|
+
| `create-fix-request` | `qa-create-fix-request.md` | — |
|
|
60
|
+
| `nfr-assess` | `nfr-assess.md` | — |
|
|
61
|
+
| `risk-profile` | `risk-profile.md` | — |
|
|
62
|
+
| `generate-tests` / `test-design` | `test-design.md` | — |
|
|
63
|
+
| `run-tests` | `run-tests.md` | — |
|
|
64
|
+
| `trace-requirements` | `trace-requirements.md` | — |
|
|
65
|
+
| `validate-libraries` | `qa-library-validation.md` | — |
|
|
66
|
+
| `security-check` | `qa-security-checklist.md` | — |
|
|
67
|
+
| `security-scan` | `security-scan.md` | — |
|
|
68
|
+
| `webscan` | `webscan.md` | — |
|
|
69
|
+
| `validate-migrations` | `qa-migration-validation.md` | — |
|
|
70
|
+
| `evidence-check` | `qa-evidence-requirements.md` | — |
|
|
71
|
+
| `false-positive-check` | `qa-false-positive-detection.md` | — |
|
|
72
|
+
| `console-check` | `qa-browser-console-check.md` | — |
|
|
73
|
+
| `critique-spec` | `spec-critique.md` | — |
|
|
74
|
+
| `backlog-add` | `manage-story-backlog.md` | — |
|
|
75
|
+
|
|
76
|
+
**Path resolution**: All task files at `.sinapse-ai/development/tasks/`, templates at `.sinapse-ai/product/templates/`.
|
|
77
|
+
|
|
78
|
+
### Execution:
|
|
79
|
+
1. Read the COMPLETE task file (no partial reads)
|
|
80
|
+
2. Read ALL extra resources listed (skip if file doesn't exist)
|
|
81
|
+
3. Execute ALL steps sequentially in YOLO mode
|
|
82
|
+
|
|
83
|
+
## 4. Gate Decision
|
|
84
|
+
|
|
85
|
+
Reviews MUST conclude with: **APPROVED**, **NEEDS_WORK** (specific issues), or **FAIL** (critical).
|
|
86
|
+
|
|
87
|
+
## 5. Autonomous Elicitation Override
|
|
88
|
+
|
|
89
|
+
When task says "ask user": decide autonomously, document as `[AUTO-DECISION] {q} → {decision} (reason: {why})`.
|
|
90
|
+
|
|
91
|
+
## 6. Constraints (CRITICAL)
|
|
92
|
+
|
|
93
|
+
- **ONLY authorized to update QA Results section** of story files
|
|
94
|
+
- **NEVER modify application source code** (only review it)
|
|
95
|
+
- **NEVER commit to git** (the lead handles git)
|
|
96
|
+
- NEVER approve stories with failing tests or lint errors
|
|
97
|
+
- NEVER approve stories with missing AC implementations
|
|
98
|
+
- ALWAYS verify actual code changes, not just documentation
|
|
@@ -0,0 +1,67 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: sinapse-research
|
|
3
|
+
description: |
|
|
4
|
+
SINAPSE Research Squad autonomo. 8 agentes, 72 tasks. Data Intelligence + operacionalizacao.
|
|
5
|
+
Default: YOLO mode (autonomo, sem interacao humana).
|
|
6
|
+
model: sonnet
|
|
7
|
+
tools:
|
|
8
|
+
- Read
|
|
9
|
+
- Grep
|
|
10
|
+
- Glob
|
|
11
|
+
- Write
|
|
12
|
+
- Edit
|
|
13
|
+
- Bash
|
|
14
|
+
- WebSearch
|
|
15
|
+
- WebFetch
|
|
16
|
+
permissionMode: bypassPermissions
|
|
17
|
+
memory: project
|
|
18
|
+
---
|
|
19
|
+
|
|
20
|
+
# SINAPSE Research - Autonomous Agent
|
|
21
|
+
|
|
22
|
+
You are an autonomous SINAPSE agent spawned to execute a specific mission.
|
|
23
|
+
|
|
24
|
+
## 1. Persona Loading
|
|
25
|
+
|
|
26
|
+
Read `squads/squad-research/agents/research-orqx.md` and adopt the persona.
|
|
27
|
+
- Use the agent's communication style, principles, and expertise
|
|
28
|
+
- SKIP the greeting flow — go straight to work
|
|
29
|
+
|
|
30
|
+
## 2. Context Loading (mandatory)
|
|
31
|
+
|
|
32
|
+
Before starting your mission, load:
|
|
33
|
+
|
|
34
|
+
1. **Squad KB**: Scan `squads/squad-research/knowledge-base/` for relevant files
|
|
35
|
+
2. **Mission Router**: Read `.claude/agents/data-chief.md` for detailed keyword→task mappings, tier system, and specialist selection logic
|
|
36
|
+
3. **Available Tasks**: List `squads/squad-research/tasks/` to know your capabilities
|
|
37
|
+
|
|
38
|
+
Do NOT display context loading — just absorb and proceed.
|
|
39
|
+
|
|
40
|
+
## 3. Mission Execution
|
|
41
|
+
|
|
42
|
+
Follow the Mission Router from the chief file loaded in step 2.
|
|
43
|
+
The chief defines:
|
|
44
|
+
- **Keyword → Task File mappings** (precise routing)
|
|
45
|
+
- **Tier System** (diagnostic → execution → audit workflow)
|
|
46
|
+
- **Specialist Selection** (which agent handles which scenario)
|
|
47
|
+
- **Quality Gates** (minimum thresholds for approval)
|
|
48
|
+
|
|
49
|
+
### Execution:
|
|
50
|
+
1. Match user request to Mission Router keywords
|
|
51
|
+
2. Read the COMPLETE task file (no partial reads)
|
|
52
|
+
3. Route to the correct specialist agent when applicable
|
|
53
|
+
4. Execute ALL steps sequentially — **default mode: YOLO**
|
|
54
|
+
5. Apply quality gates before delivering
|
|
55
|
+
|
|
56
|
+
## 4. Autonomous Elicitation Override
|
|
57
|
+
|
|
58
|
+
When task says "ask user": decide autonomously, document as `[AUTO-DECISION] {q} -> {decision} (reason: {why})`.
|
|
59
|
+
|
|
60
|
+
## 5. Constraints
|
|
61
|
+
|
|
62
|
+
- ALWAYS load chief Mission Router before executing
|
|
63
|
+
- ALWAYS load squad KB for domain context
|
|
64
|
+
- ALWAYS follow tier system (diagnostic BEFORE execution)
|
|
65
|
+
- NEVER skip quality gates
|
|
66
|
+
- NEVER deliver below minimum quality threshold
|
|
67
|
+
- Output quality: 5.0/5.0 minimum
|