conectese 0.1.15 → 0.2.0
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.
- package/README.md +2 -2
- package/agents/conectese_admin/AGENT.md +10 -5
- package/agents/data-extractor/AGENT.md +35 -3
- package/agents/direito-administrativo/AGENT.md +68 -22
- package/agents/direito-aduaneiro/AGENT.md +55 -0
- package/agents/direito-aeroportuario/AGENT.md +53 -0
- package/agents/direito-agrario/AGENT.md +73 -0
- package/agents/direito-ambiental/AGENT.md +61 -24
- package/agents/direito-bancario/AGENT.md +79 -0
- package/agents/direito-civil/AGENT.md +71 -20
- package/agents/direito-constitucional/AGENT.md +79 -0
- package/agents/direito-da-crianca-e-do-adolescente-eca/AGENT.md +69 -0
- package/agents/direito-da-propriedade-intelectual/AGENT.md +65 -0
- package/agents/direito-de-familia/AGENT.md +92 -0
- package/agents/direito-de-transito/AGENT.md +59 -0
- package/agents/direito-desportivo/AGENT.md +40 -27
- package/agents/direito-digital/AGENT.md +61 -24
- package/agents/direito-do-consumidor/AGENT.md +77 -0
- package/agents/direito-do-trabalho/AGENT.md +73 -20
- package/agents/direito-economico/AGENT.md +58 -0
- package/agents/direito-eleitoral/AGENT.md +56 -24
- package/agents/direito-empresarial/AGENT.md +62 -24
- package/agents/direito-financeiro/AGENT.md +63 -0
- package/agents/direito-imobiliario/AGENT.md +79 -0
- package/agents/direito-indigena/AGENT.md +72 -0
- package/agents/direito-internacional/AGENT.md +55 -24
- package/agents/direito-maritimo/AGENT.md +55 -0
- package/agents/direito-medico-e-da-saude/AGENT.md +69 -0
- package/agents/direito-militar/AGENT.md +54 -24
- package/agents/direito-notarial-e-registral/AGENT.md +47 -24
- package/agents/direito-penal/AGENT.md +74 -19
- package/agents/direito-previdenciario/AGENT.md +84 -0
- package/agents/direito-processal-civil/AGENT.md +72 -20
- package/agents/direito-processal-do-trabalho/AGENT.md +58 -24
- package/agents/direito-processal-militar/AGENT.md +38 -27
- package/agents/direito-processal-penal/AGENT.md +62 -24
- package/agents/direito-processual-civil/AGENT.md +94 -0
- package/agents/direito-processual-do-trabalho/AGENT.md +76 -0
- package/agents/direito-processual-militar/AGENT.md +53 -0
- package/agents/direito-processual-penal/AGENT.md +80 -0
- package/agents/direito-securitario/AGENT.md +75 -0
- package/agents/direito-sindical/AGENT.md +46 -24
- package/agents/direito-societario/AGENT.md +65 -0
- package/agents/direito-tributario/AGENT.md +98 -0
- package/agents/direito-urbanistico/AGENT.md +58 -0
- package/agents/direitos-humanos/AGENT.md +76 -0
- package/agents/legal-analyst/AGENT.md +1 -1
- package/agents/legal-designer/AGENT.md +8 -3
- package/agents/lgpd-anonymizer/AGENT.md +66 -4
- package/agents/lgpd-restorer/AGENT.md +38 -4
- package/agents/task-router/AGENT.md +47 -3
- package/bin/conectese.js +4 -4
- package/dashboard/src/App.tsx +6 -4
- package/dashboard/src/components/CheckpointModal.tsx +150 -0
- package/dashboard/src/components/StatusBadge.tsx +3 -3
- package/dashboard/src/components/StatusBar.tsx +7 -7
- package/dashboard/src/components/{SquadCard.tsx → TeamCard.tsx} +7 -7
- package/dashboard/src/components/{SquadSelector.tsx → TeamSelector.tsx} +18 -18
- package/dashboard/src/hooks/{useSquadSocket.ts → useTeamSocket.ts} +13 -13
- package/dashboard/src/lib/normalizeState.ts +3 -3
- package/dashboard/src/office/OfficeScene.ts +3 -3
- package/dashboard/src/office/PhaserGame.tsx +6 -6
- package/dashboard/src/plugin/{squadWatcher.ts → teamWatcher.ts} +83 -44
- package/dashboard/src/store/useTeamStore.ts +56 -0
- package/dashboard/src/types/state.ts +10 -9
- package/dashboard/tsconfig.tsbuildinfo +1 -1
- package/dashboard/vite.config.ts +2 -2
- package/package.json +12 -9
- package/skills/README.md +1 -1
- package/skills/conectese-agent-creator/SKILL.md +1 -1
- package/skills/conectese-design/SKILL.md +50 -0
- package/skills/conectese-skill-creator/SKILL.md +3 -3
- package/skills/conectese-skill-creator/references/skill-format.md +4 -4
- package/skills/document-analyzer/SKILL.md +40 -0
- package/skills/document-analyzer/scripts/parse-doc.js +113 -0
- package/skills/image-ai-generator/SKILL.md +5 -5
- package/skills/image-creator/SKILL.md +3 -3
- package/skills/image-fetcher/SKILL.md +5 -5
- package/skills/instagram-publisher/SKILL.md +2 -2
- package/skills/jurisprudencia-validator/SKILL.md +45 -0
- package/skills/jurisprudencia-validator/scripts/validate-urls.js +103 -0
- package/skills/legal-pricing/SKILL.md +33 -0
- package/skills/legal-pricing/scripts/pricing-engine.js +72 -0
- package/skills/resend/SKILL.md +2 -2
- package/skills/template-designer/SKILL.md +29 -29
- package/src/agents-cli.js +1 -1
- package/src/agents.js +1 -1
- package/src/init.js +2 -1
- package/src/locales/en.json +2 -2
- package/src/locales/es.json +2 -2
- package/src/locales/pt-BR.json +2 -2
- package/src/readme/README.md +26 -26
- package/src/runs.js +14 -14
- package/src/skills-cli.js +1 -1
- package/src/update.js +1 -1
- package/templates/_conectese/.conectese-version +1 -1
- package/templates/ide-templates/antigravity/.agent/rules/conectese.md +10 -10
- package/templates/ide-templates/antigravity/.agent/workflows/conectese.md +18 -18
- package/templates/ide-templates/claude-code/.claude/skills/conectese/SKILL.md +31 -31
- package/templates/ide-templates/claude-code/CLAUDE.md +10 -10
- package/templates/ide-templates/codex/.agents/skills/conectese/SKILL.md +1 -1
- package/templates/ide-templates/codex/AGENTS.md +18 -18
- package/templates/ide-templates/cursor/.cursor/commands/conectese.md +1 -1
- package/templates/ide-templates/opencode/.opencode/commands/conectese.md +1 -1
- package/templates/ide-templates/opencode/AGENTS.md +18 -18
- package/templates/ide-templates/vscode-copilot/.github/prompts/conectese.prompt.md +33 -33
- package/_conectese/.conectese-version +0 -1
- package/_conectese/_memory/company/modelo_contrato.md +0 -18
- package/_conectese/_memory/company/modelo_procuracao.md +0 -15
- package/_conectese/_memory/company/modelo_recibo.md +0 -13
- package/_conectese/config/playwright.config.json +0 -11
- package/_conectese/core/architect.agent.yaml +0 -110
- package/_conectese/core/best-practices/_catalog.yaml +0 -116
- package/_conectese/core/best-practices/blog-post.md +0 -132
- package/_conectese/core/best-practices/blog-seo.md +0 -127
- package/_conectese/core/best-practices/copywriting.md +0 -426
- package/_conectese/core/best-practices/data-analysis.md +0 -401
- package/_conectese/core/best-practices/email-newsletter.md +0 -118
- package/_conectese/core/best-practices/email-sales.md +0 -110
- package/_conectese/core/best-practices/image-design.md +0 -348
- package/_conectese/core/best-practices/instagram-feed.md +0 -235
- package/_conectese/core/best-practices/instagram-reels.md +0 -112
- package/_conectese/core/best-practices/instagram-stories.md +0 -107
- package/_conectese/core/best-practices/linkedin-article.md +0 -116
- package/_conectese/core/best-practices/linkedin-post.md +0 -121
- package/_conectese/core/best-practices/researching.md +0 -349
- package/_conectese/core/best-practices/review.md +0 -269
- package/_conectese/core/best-practices/social-networks-publishing.md +0 -294
- package/_conectese/core/best-practices/strategist.md +0 -344
- package/_conectese/core/best-practices/technical-writing.md +0 -365
- package/_conectese/core/best-practices/twitter-post.md +0 -105
- package/_conectese/core/best-practices/twitter-thread.md +0 -122
- package/_conectese/core/best-practices/whatsapp-broadcast.md +0 -107
- package/_conectese/core/best-practices/youtube-script.md +0 -122
- package/_conectese/core/best-practices/youtube-shorts.md +0 -112
- package/_conectese/core/prompts/build.prompt.md +0 -547
- package/_conectese/core/prompts/design.prompt.md +0 -469
- package/_conectese/core/prompts/discovery.prompt.md +0 -269
- package/_conectese/core/prompts/sherlock-instagram.md +0 -123
- package/_conectese/core/prompts/sherlock-linkedin.md +0 -73
- package/_conectese/core/prompts/sherlock-shared.md +0 -684
- package/_conectese/core/prompts/sherlock-twitter.md +0 -78
- package/_conectese/core/prompts/sherlock-youtube.md +0 -85
- package/_conectese/core/runner.pipeline.md +0 -535
- package/_conectese/core/skills.engine.md +0 -381
- package/agents/direito-adaneiro/AGENT.md +0 -42
- package/agents/direito-aeroporta-rio/AGENT.md +0 -42
- package/agents/direito-agra-rio/AGENT.md +0 -42
- package/agents/direito-banca-rio/AGENT.md +0 -42
- package/agents/direito-constitcional/AGENT.md +0 -42
- package/agents/direito-da-crianc-a-e-do-adolescente-eca/AGENT.md +0 -42
- package/agents/direito-da-propriedade-intelectal/AGENT.md +0 -42
- package/agents/direito-de-ami-lia/AGENT.md +0 -42
- package/agents/direito-de-tra-nsito/AGENT.md +0 -42
- package/agents/direito-do-consmidor/AGENT.md +0 -42
- package/agents/direito-econo-mico/AGENT.md +0 -42
- package/agents/direito-imobilia-rio/AGENT.md +0 -42
- package/agents/direito-inanceiro/AGENT.md +0 -42
- package/agents/direito-mari-timo/AGENT.md +0 -42
- package/agents/direito-me-dico-e-da-sa-de/AGENT.md +0 -42
- package/agents/direito-ndia-rio/AGENT.md +0 -42
- package/agents/direito-previdencia-rio/AGENT.md +0 -42
- package/agents/direito-rbani-stico/AGENT.md +0 -42
- package/agents/direito-secrita-rio/AGENT.md +0 -42
- package/agents/direito-societa-rio/AGENT.md +0 -42
- package/agents/direito-tribta-rio/AGENT.md +0 -42
- package/agents/direitos-hmanos/AGENT.md +0 -42
- package/dashboard/src/store/useSquadStore.ts +0 -56
package/src/locales/en.json
CHANGED
|
@@ -24,7 +24,7 @@
|
|
|
24
24
|
"updateStartingUnknown": "Updating Conectese (unknown version) → {new}...",
|
|
25
25
|
"updatedFile": "📄 Updated {path}",
|
|
26
26
|
"updateSuccess": "✅ Conectese {version} installed successfully!",
|
|
27
|
-
"updatePreserved": "✓ Preserved: _memory/, _investigations/,
|
|
27
|
+
"updatePreserved": "✓ Preserved: _memory/, _investigations/, teams/",
|
|
28
28
|
"updateFileCount": "✓ Updated: {count} system files",
|
|
29
29
|
"updateLatestHint": "💡 Tip: Use 'npx conectese@latest update' to always get the newest version.",
|
|
30
30
|
"skillsNotInitialized": "No Conectese installation found. Run 'init' first.",
|
|
@@ -69,5 +69,5 @@
|
|
|
69
69
|
"step1VsCodeCopilot": "1. Open this directory in VS Code",
|
|
70
70
|
"step2VsCodeCopilot": "2. Open Copilot Chat (Ctrl+Alt+I)",
|
|
71
71
|
"step3VsCodeCopilot": "3. Type /conectese to get started",
|
|
72
|
-
"tokenCostWarning": "Running
|
|
72
|
+
"tokenCostWarning": "Running teams consumes AI tokens (paid). Monitor your usage in your IDE or AI provider dashboard."
|
|
73
73
|
}
|
package/src/locales/es.json
CHANGED
|
@@ -23,7 +23,7 @@
|
|
|
23
23
|
"updateStartingUnknown": "Actualizando Conectese (versión desconocida) → {new}...",
|
|
24
24
|
"updatedFile": "📄 Actualizado {path}",
|
|
25
25
|
"updateSuccess": "✅ ¡Conectese {version} instalado correctamente!",
|
|
26
|
-
"updatePreserved": "✓ Preservado: _memory/, _investigations/,
|
|
26
|
+
"updatePreserved": "✓ Preservado: _memory/, _investigations/, teams/",
|
|
27
27
|
"updateFileCount": "✓ Actualizados: {count} archivos del sistema",
|
|
28
28
|
"updateLatestHint": "💡 Consejo: Use 'npx conectese@latest update' para obtener siempre la versión más reciente.",
|
|
29
29
|
"skillsNotInitialized": "Instalación de Conectese no encontrada. Ejecuta 'init' primero.",
|
|
@@ -68,5 +68,5 @@
|
|
|
68
68
|
"step1VsCodeCopilot": "1. Abre este directorio en VS Code",
|
|
69
69
|
"step2VsCodeCopilot": "2. Abre el Copilot Chat (Ctrl+Alt+I)",
|
|
70
70
|
"step3VsCodeCopilot": "3. Escribe /conectese para empezar",
|
|
71
|
-
"tokenCostWarning": "Ejecutar
|
|
71
|
+
"tokenCostWarning": "Ejecutar teams consume tokens de IA (de pago). Monitorea tu consumo en tu IDE o en el dashboard de tu proveedor de IA."
|
|
72
72
|
}
|
package/src/locales/pt-BR.json
CHANGED
|
@@ -23,7 +23,7 @@
|
|
|
23
23
|
"updateStartingUnknown": "Atualizando Conectese (versão desconhecida) → {new}...",
|
|
24
24
|
"updatedFile": "📄 Atualizado {path}",
|
|
25
25
|
"updateSuccess": "✅ Conectese {version} instalado com sucesso!",
|
|
26
|
-
"updatePreserved": "✓ Preservado: _memory/, _investigations/,
|
|
26
|
+
"updatePreserved": "✓ Preservado: _memory/, _investigations/, teams/",
|
|
27
27
|
"updateFileCount": "✓ Atualizados: {count} arquivos do sistema",
|
|
28
28
|
"updateLatestHint": "💡 Dica: Use 'npx conectese@latest update' para sempre obter a versão mais recente.",
|
|
29
29
|
"skillsNotInitialized": "Instalação Conectese não encontrada. Execute 'init' primeiro.",
|
|
@@ -68,5 +68,5 @@
|
|
|
68
68
|
"step1VsCodeCopilot": "1. Abra este diretório no VS Code",
|
|
69
69
|
"step2VsCodeCopilot": "2. Abra o Copilot Chat (Ctrl+Alt+I)",
|
|
70
70
|
"step3VsCodeCopilot": "3. Digite /conectese para começar",
|
|
71
|
-
"tokenCostWarning": "Executar
|
|
71
|
+
"tokenCostWarning": "Executar times consome tokens de IA (pagos). Monitore seu consumo na sua IDE ou no dashboard do provedor de IA."
|
|
72
72
|
}
|
package/src/readme/README.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
# Conectese
|
|
2
2
|
|
|
3
|
-
Crie
|
|
3
|
+
Crie times de agentes de IA que trabalham juntos — direto do seu IDE.
|
|
4
4
|
|
|
5
5
|
## Como Usar
|
|
6
6
|
|
|
@@ -10,34 +10,34 @@ Abra esta pasta no seu IDE e digite:
|
|
|
10
10
|
/conectese
|
|
11
11
|
```
|
|
12
12
|
|
|
13
|
-
Isso abre o menu principal. De lá você pode criar
|
|
13
|
+
Isso abre o menu principal. De lá você pode criar times, executá-los e mais.
|
|
14
14
|
|
|
15
15
|
Você também pode ser direto — descreva o que quer em linguagem natural:
|
|
16
16
|
|
|
17
17
|
```
|
|
18
|
-
/conectese crie um
|
|
19
|
-
/conectese execute o
|
|
18
|
+
/conectese crie um time para escrever posts no LinkedIn sobre IA
|
|
19
|
+
/conectese execute o time meu-time
|
|
20
20
|
```
|
|
21
21
|
|
|
22
|
-
## Criar um
|
|
22
|
+
## Criar um Time
|
|
23
23
|
|
|
24
|
-
Digite `/conectese` e escolha "Criar
|
|
24
|
+
Digite `/conectese` e escolha "Criar time" no menu, ou seja direto:
|
|
25
25
|
|
|
26
26
|
```
|
|
27
|
-
/conectese crie um
|
|
27
|
+
/conectese crie um time para [o que você precisa]
|
|
28
28
|
```
|
|
29
29
|
|
|
30
|
-
O Arquiteto fará algumas perguntas, projetará o
|
|
30
|
+
O Arquiteto fará algumas perguntas, projetará o time e configurará tudo automaticamente.
|
|
31
31
|
|
|
32
|
-
## Executar um
|
|
32
|
+
## Executar um Time
|
|
33
33
|
|
|
34
|
-
Digite `/conectese` e escolha "Executar
|
|
34
|
+
Digite `/conectese` e escolha "Executar time" no menu, ou seja direto:
|
|
35
35
|
|
|
36
36
|
```
|
|
37
|
-
/conectese execute o
|
|
37
|
+
/conectese execute o time <nome-do-time>
|
|
38
38
|
```
|
|
39
39
|
|
|
40
|
-
O
|
|
40
|
+
O time executa automaticamente, pausando apenas nos checkpoints de decisão.
|
|
41
41
|
|
|
42
42
|
## Escritório Virtual
|
|
43
43
|
|
|
@@ -52,7 +52,7 @@ O Escritório Virtual é uma interface visual 2D que mostra seus agentes trabalh
|
|
|
52
52
|
**Passo 2 — Sirva localmente** (no terminal):
|
|
53
53
|
|
|
54
54
|
```bash
|
|
55
|
-
npx serve
|
|
55
|
+
npx serve teams/<nome-do-time>/dashboard
|
|
56
56
|
```
|
|
57
57
|
|
|
58
58
|
**Passo 3 —** Abra `http://localhost:3000` no seu navegador.
|
|
@@ -61,7 +61,7 @@ npx serve squads/<nome-do-squad>/dashboard
|
|
|
61
61
|
|
|
62
62
|
# Conectese (English)
|
|
63
63
|
|
|
64
|
-
Create AI
|
|
64
|
+
Create AI teams that work together — right from your IDE.
|
|
65
65
|
|
|
66
66
|
## How to Use
|
|
67
67
|
|
|
@@ -71,34 +71,34 @@ Open this folder in your IDE and type:
|
|
|
71
71
|
/conectese
|
|
72
72
|
```
|
|
73
73
|
|
|
74
|
-
This opens the main menu. From there you can create
|
|
74
|
+
This opens the main menu. From there you can create teams, run them, and more.
|
|
75
75
|
|
|
76
76
|
You can also be direct — describe what you want in plain language:
|
|
77
77
|
|
|
78
78
|
```
|
|
79
|
-
/conectese create a
|
|
80
|
-
/conectese run my-
|
|
79
|
+
/conectese create a team for writing LinkedIn posts about AI
|
|
80
|
+
/conectese run my-team
|
|
81
81
|
```
|
|
82
82
|
|
|
83
|
-
## Create a
|
|
83
|
+
## Create a Team
|
|
84
84
|
|
|
85
|
-
Type `/conectese` and choose "Create
|
|
85
|
+
Type `/conectese` and choose "Create team" from the menu, or be direct:
|
|
86
86
|
|
|
87
87
|
```
|
|
88
|
-
/conectese create a
|
|
88
|
+
/conectese create a team for [what you need]
|
|
89
89
|
```
|
|
90
90
|
|
|
91
|
-
The Architect will ask a few questions, design the
|
|
91
|
+
The Architect will ask a few questions, design the team, and set everything up automatically.
|
|
92
92
|
|
|
93
|
-
## Run a
|
|
93
|
+
## Run a Team
|
|
94
94
|
|
|
95
|
-
Type `/conectese` and choose "Run
|
|
95
|
+
Type `/conectese` and choose "Run team" from the menu, or be direct:
|
|
96
96
|
|
|
97
97
|
```
|
|
98
|
-
/conectese run the <
|
|
98
|
+
/conectese run the <team-name> team
|
|
99
99
|
```
|
|
100
100
|
|
|
101
|
-
The
|
|
101
|
+
The team runs automatically, pausing only at decision checkpoints.
|
|
102
102
|
|
|
103
103
|
## Virtual Office
|
|
104
104
|
|
|
@@ -113,7 +113,7 @@ The Virtual Office is a 2D visual interface that shows your agents working in re
|
|
|
113
113
|
**Step 2 — Serve it locally** (in terminal):
|
|
114
114
|
|
|
115
115
|
```bash
|
|
116
|
-
npx serve
|
|
116
|
+
npx serve teams/<team-name>/dashboard
|
|
117
117
|
```
|
|
118
118
|
|
|
119
119
|
**Step 3 —** Open `http://localhost:3000` in your browser.
|
package/src/runs.js
CHANGED
|
@@ -3,16 +3,16 @@ import { join } from 'node:path';
|
|
|
3
3
|
|
|
4
4
|
const MAX_RUNS = 20;
|
|
5
5
|
|
|
6
|
-
export async function listRuns(
|
|
7
|
-
const
|
|
8
|
-
let
|
|
6
|
+
export async function listRuns(teamName, targetDir = process.cwd()) {
|
|
7
|
+
const teamsDir = join(targetDir, 'teams');
|
|
8
|
+
let teamNames;
|
|
9
9
|
|
|
10
10
|
try {
|
|
11
|
-
if (
|
|
12
|
-
|
|
11
|
+
if (teamName) {
|
|
12
|
+
teamNames = [teamName];
|
|
13
13
|
} else {
|
|
14
|
-
const entries = await readdir(
|
|
15
|
-
|
|
14
|
+
const entries = await readdir(teamsDir, { withFileTypes: true });
|
|
15
|
+
teamNames = entries.filter((e) => e.isDirectory()).map((e) => e.name);
|
|
16
16
|
}
|
|
17
17
|
} catch {
|
|
18
18
|
return [];
|
|
@@ -20,8 +20,8 @@ export async function listRuns(squadName, targetDir = process.cwd()) {
|
|
|
20
20
|
|
|
21
21
|
const runs = [];
|
|
22
22
|
|
|
23
|
-
for (const name of
|
|
24
|
-
const outputDir = join(
|
|
23
|
+
for (const name of teamNames) {
|
|
24
|
+
const outputDir = join(teamsDir, name, 'output');
|
|
25
25
|
let runDirs;
|
|
26
26
|
try {
|
|
27
27
|
const entries = await readdir(outputDir, { withFileTypes: true });
|
|
@@ -31,7 +31,7 @@ export async function listRuns(squadName, targetDir = process.cwd()) {
|
|
|
31
31
|
}
|
|
32
32
|
|
|
33
33
|
for (const runId of runDirs) {
|
|
34
|
-
const run = {
|
|
34
|
+
const run = { team: name, runId, status: 'unknown', steps: null, duration: null };
|
|
35
35
|
|
|
36
36
|
try {
|
|
37
37
|
const raw = await readFile(join(outputDir, runId, 'state.json'), 'utf-8');
|
|
@@ -73,11 +73,11 @@ export function printRuns(runs) {
|
|
|
73
73
|
return;
|
|
74
74
|
}
|
|
75
75
|
|
|
76
|
-
let
|
|
76
|
+
let currentTeam = null;
|
|
77
77
|
for (const run of runs) {
|
|
78
|
-
if (run.
|
|
79
|
-
|
|
80
|
-
console.log(`\n ${
|
|
78
|
+
if (run.team !== currentTeam) {
|
|
79
|
+
currentTeam = run.team;
|
|
80
|
+
console.log(`\n ${currentTeam}`);
|
|
81
81
|
console.log(' ' + '─'.repeat(50));
|
|
82
82
|
}
|
|
83
83
|
const parts = [` ${run.runId}`];
|
package/src/skills-cli.js
CHANGED
|
@@ -76,7 +76,7 @@ async function runList(targetDir) {
|
|
|
76
76
|
console.log(` ${t('skillsNoneInstalled')}`);
|
|
77
77
|
}
|
|
78
78
|
|
|
79
|
-
console.log(`\n Browse available skills at: https://github.com/
|
|
79
|
+
console.log(`\n Browse available skills at: https://github.com/henriqueadm-ai/conectese/tree/main/skills\n`);
|
|
80
80
|
}
|
|
81
81
|
|
|
82
82
|
async function runInstall(id, targetDir) {
|
package/src/update.js
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
0.
|
|
1
|
+
0.2.0
|
|
@@ -9,8 +9,8 @@ This project uses **Conectese**, a multi-agent orchestration framework.
|
|
|
9
9
|
## Quick Start
|
|
10
10
|
|
|
11
11
|
Type `/conectese` to open the main menu, or use any of these commands:
|
|
12
|
-
- `/conectese create` — Create a new
|
|
13
|
-
- `/conectese run <name>` — Run a
|
|
12
|
+
- `/conectese create` — Create a new team
|
|
13
|
+
- `/conectese run <name>` — Run a team
|
|
14
14
|
- `/conectese help` — See all commands
|
|
15
15
|
|
|
16
16
|
## Directory Structure
|
|
@@ -18,17 +18,17 @@ Type `/conectese` to open the main menu, or use any of these commands:
|
|
|
18
18
|
- `_conectese/` — Conectese core files (do not modify manually)
|
|
19
19
|
- `_conectese/_memory/` — Persistent memory (company context, preferences)
|
|
20
20
|
- `skills/` — Installed skills (integrations, scripts, prompts)
|
|
21
|
-
- `
|
|
22
|
-
- `
|
|
23
|
-
- `
|
|
21
|
+
- `teams/` — User-created teams
|
|
22
|
+
- `teams/{name}/_investigations/` — Sherlock content investigations (profile analyses)
|
|
23
|
+
- `teams/{name}/output/` — Generated content and files
|
|
24
24
|
- `_conectese/_browser_profile/` — Persistent browser sessions (login cookies, localStorage)
|
|
25
25
|
|
|
26
26
|
## How It Works
|
|
27
27
|
|
|
28
28
|
1. The `/conectese` workflow is the entry point for all interactions
|
|
29
|
-
2. The **Architect** agent creates and modifies
|
|
30
|
-
3. During
|
|
31
|
-
4. The **Pipeline Runner** executes
|
|
29
|
+
2. The **Architect** agent creates and modifies teams
|
|
30
|
+
3. During team creation, the **Sherlock** investigator can analyze reference profiles (Instagram, YouTube, Twitter/X, LinkedIn) to extract real content patterns
|
|
31
|
+
4. The **Pipeline Runner** executes teams automatically
|
|
32
32
|
5. All tasks run inline and sequentially (no background subagents)
|
|
33
33
|
6. Checkpoints pause execution for user input/approval
|
|
34
34
|
|
|
@@ -36,8 +36,8 @@ Type `/conectese` to open the main menu, or use any of these commands:
|
|
|
36
36
|
|
|
37
37
|
- Always use `/conectese` commands to interact with the system
|
|
38
38
|
- Do not manually edit files in `_conectese/core/` unless you know what you're doing
|
|
39
|
-
-
|
|
40
|
-
- Company context in `_conectese/_memory/company.md` is loaded for every
|
|
39
|
+
- Team YAML files can be edited manually if needed, but prefer using `/conectese edit`
|
|
40
|
+
- Company context in `_conectese/_memory/company.md` is loaded for every team run
|
|
41
41
|
|
|
42
42
|
## Antigravity Environment: Subagents
|
|
43
43
|
|
|
@@ -1,5 +1,5 @@
|
|
|
1
1
|
---
|
|
2
|
-
description: Conectese — Create and run AI agent
|
|
2
|
+
description: Conectese — Create and run AI agent teams for your business
|
|
3
3
|
---
|
|
4
4
|
|
|
5
5
|
You are now activating the Conectese system. Follow these steps IN ORDER:
|
|
@@ -32,13 +32,13 @@ If `company.md` is empty or contains `<!-- NOT CONFIGURED -->`:
|
|
|
32
32
|
Present the following numbered menu and ask the user to reply with a number:
|
|
33
33
|
|
|
34
34
|
**Primary menu:**
|
|
35
|
-
1. **Create a new
|
|
36
|
-
2. **Run an existing
|
|
37
|
-
3. **My
|
|
35
|
+
1. **Create a new team** — Describe what you need and I'll build a team for you
|
|
36
|
+
2. **Run an existing team** — Execute a team's pipeline
|
|
37
|
+
3. **My teams** — View, edit, or delete your teams
|
|
38
38
|
4. **More options** — Skills, company profile, settings, and help
|
|
39
39
|
|
|
40
40
|
If the user replies "4" or types "More options", present a second numbered menu:
|
|
41
|
-
1. **Skills** — Browse, install, create, and manage skills for your
|
|
41
|
+
1. **Skills** — Browse, install, create, and manage skills for your teams
|
|
42
42
|
2. **Company profile** — View or update your company information
|
|
43
43
|
3. **Settings & Help** — Language, preferences, configuration, and help
|
|
44
44
|
|
|
@@ -50,19 +50,19 @@ Parse user input and route to the appropriate action:
|
|
|
50
50
|
|---------------|--------|
|
|
51
51
|
| `/conectese` or `/conectese menu` | Show main menu |
|
|
52
52
|
| `/conectese help` | Show help text |
|
|
53
|
-
| `/conectese create <description>` | Load Architect → Create
|
|
54
|
-
| `/conectese list` | List all
|
|
55
|
-
| `/conectese run <name>` | Load Pipeline Runner → Execute
|
|
56
|
-
| `/conectese edit <name> <changes>` | Load Architect → Edit
|
|
53
|
+
| `/conectese create <description>` | Load Architect → Create Team flow |
|
|
54
|
+
| `/conectese list` | List all teams in `teams/` directory |
|
|
55
|
+
| `/conectese run <name>` | Load Pipeline Runner → Execute team |
|
|
56
|
+
| `/conectese edit <name> <changes>` | Load Architect → Edit Team flow |
|
|
57
57
|
| `/conectese skills` | Load Skills Engine → Show skills menu |
|
|
58
58
|
| `/conectese install <name>` | Install a skill from the catalog |
|
|
59
59
|
| `/conectese uninstall <name>` | Remove an installed skill |
|
|
60
|
-
| `/conectese delete <name>` | Confirm and delete
|
|
60
|
+
| `/conectese delete <name>` | Confirm and delete team directory |
|
|
61
61
|
| `/conectese edit-company` | Re-run company profile setup |
|
|
62
62
|
| `/conectese show-company` | Display company.md contents |
|
|
63
63
|
| `/conectese settings` | Show/edit preferences.md |
|
|
64
64
|
| `/conectese reset` | Confirm and reset all configuration |
|
|
65
|
-
| Natural language about
|
|
65
|
+
| Natural language about teams | Infer intent and route accordingly |
|
|
66
66
|
|
|
67
67
|
## Loading Agents
|
|
68
68
|
|
|
@@ -75,13 +75,13 @@ When a specific agent needs to be activated:
|
|
|
75
75
|
|
|
76
76
|
## Loading the Pipeline Runner
|
|
77
77
|
|
|
78
|
-
When running a
|
|
78
|
+
When running a team:
|
|
79
79
|
|
|
80
|
-
1. Read `
|
|
81
|
-
2. Read `
|
|
80
|
+
1. Read `teams/{name}/team.yaml` to understand the pipeline
|
|
81
|
+
2. Read `teams/{name}/team-party.csv` to load all agent personas
|
|
82
82
|
3. For each agent in the party CSV, also read their full `.agent.md` file from agents/ directory
|
|
83
83
|
4. Load company context from `_conectese/_memory/company.md`
|
|
84
|
-
5. Load
|
|
84
|
+
5. Load team memory from `teams/{name}/_memory/memories.md`
|
|
85
85
|
6. Read the pipeline runner instructions from `_conectese/core/runner.pipeline.md`
|
|
86
86
|
7. Execute the pipeline step by step following runner instructions
|
|
87
87
|
|
|
@@ -95,8 +95,8 @@ When running a squad:
|
|
|
95
95
|
## Critical Rules
|
|
96
96
|
|
|
97
97
|
- NEVER skip the onboarding if company.md is not configured
|
|
98
|
-
- ALWAYS load company context before running any
|
|
98
|
+
- ALWAYS load company context before running any team
|
|
99
99
|
- ALWAYS present checkpoints to the user — never skip them
|
|
100
|
-
- ALWAYS save outputs to the
|
|
100
|
+
- ALWAYS save outputs to the team's output directory
|
|
101
101
|
- When switching personas (inline execution), clearly indicate which agent is speaking
|
|
102
|
-
- After each pipeline run, update the
|
|
102
|
+
- After each pipeline run, update the team's memories.md with key learnings
|
|
@@ -1,11 +1,11 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: conectese
|
|
3
|
-
description: "Conectese — Multi-agent orchestration framework. Create and run AI
|
|
3
|
+
description: "Conectese — Multi-agent orchestration framework. Create and run AI teams for your business."
|
|
4
4
|
---
|
|
5
5
|
|
|
6
6
|
# Conectese — Multi-Agent Orchestration
|
|
7
7
|
|
|
8
|
-
You are now operating as the Conectese system. Your primary role is to help users create, manage, and run AI agent
|
|
8
|
+
You are now operating as the Conectese system. Your primary role is to help users create, manage, and run AI agent teams.
|
|
9
9
|
|
|
10
10
|
## Initialization
|
|
11
11
|
|
|
@@ -39,13 +39,13 @@ If `company.md` is empty or contains `<!-- NOT CONFIGURED -->`:
|
|
|
39
39
|
When the user types `/conectese` or asks for the menu, present an interactive selector using AskUserQuestion with these options (max 4 per question):
|
|
40
40
|
|
|
41
41
|
**Primary menu (first question):**
|
|
42
|
-
- **Create a new
|
|
43
|
-
- **Run an existing
|
|
44
|
-
- **My
|
|
42
|
+
- **Create a new team** — Describe what you need and I'll build a team for you
|
|
43
|
+
- **Run an existing team** — Execute a team's pipeline
|
|
44
|
+
- **My teams** — View, edit, or delete your teams
|
|
45
45
|
- **More options** — Skills, company profile, settings, and help
|
|
46
46
|
|
|
47
47
|
If the user selects "More options", present a second AskUserQuestion:
|
|
48
|
-
- **Skills** — Browse, install, create, and manage skills for your
|
|
48
|
+
- **Skills** — Browse, install, create, and manage skills for your teams
|
|
49
49
|
- **Company profile** — View or update your company information
|
|
50
50
|
- **Settings & Help** — Language, preferences, configuration, and help
|
|
51
51
|
|
|
@@ -57,19 +57,19 @@ Parse user input and route to the appropriate action:
|
|
|
57
57
|
|---------------|--------|
|
|
58
58
|
| `/conectese` or `/conectese menu` | Show main menu |
|
|
59
59
|
| `/conectese help` | Show help text |
|
|
60
|
-
| `/conectese create <description>` | Load Architect → Create
|
|
61
|
-
| `/conectese list` | List all
|
|
62
|
-
| `/conectese run <name>` | Load Pipeline Runner → Execute
|
|
63
|
-
| `/conectese edit <name> <changes>` | Load Architect → Edit
|
|
60
|
+
| `/conectese create <description>` | Load Architect → Create Team flow (will ask for reference profile URLs for Sherlock investigation) |
|
|
61
|
+
| `/conectese list` | List all teams in `teams/` directory |
|
|
62
|
+
| `/conectese run <name>` | Load Pipeline Runner → Execute team |
|
|
63
|
+
| `/conectese edit <name> <changes>` | Load Architect → Edit Team flow |
|
|
64
64
|
| `/conectese skills` | Load Skills Engine → Show skills menu |
|
|
65
65
|
| `/conectese install <name>` | Install a skill from the catalog |
|
|
66
66
|
| `/conectese uninstall <name>` | Remove an installed skill |
|
|
67
|
-
| `/conectese delete <name>` | Confirm and delete
|
|
67
|
+
| `/conectese delete <name>` | Confirm and delete team directory |
|
|
68
68
|
| `/conectese edit-company` | Re-run company profile setup |
|
|
69
69
|
| `/conectese show-company` | Display company.md contents |
|
|
70
70
|
| `/conectese settings` | Show/edit preferences.md |
|
|
71
71
|
| `/conectese reset` | Confirm and reset all configuration |
|
|
72
|
-
| Natural language about
|
|
72
|
+
| Natural language about teams | Infer intent and route accordingly |
|
|
73
73
|
|
|
74
74
|
## Help Text
|
|
75
75
|
|
|
@@ -84,12 +84,12 @@ GETTING STARTED
|
|
|
84
84
|
/conectese Open the main menu
|
|
85
85
|
/conectese help Show this help
|
|
86
86
|
|
|
87
|
-
|
|
88
|
-
/conectese create Create a new
|
|
89
|
-
/conectese list List all your
|
|
90
|
-
/conectese run <name> Run a
|
|
91
|
-
/conectese edit <name> Modify an existing
|
|
92
|
-
/conectese delete <name> Delete a
|
|
87
|
+
TEAMS
|
|
88
|
+
/conectese create Create a new team (describe what you need)
|
|
89
|
+
/conectese list List all your teams
|
|
90
|
+
/conectese run <name> Run a team's pipeline
|
|
91
|
+
/conectese edit <name> Modify an existing team
|
|
92
|
+
/conectese delete <name> Delete a team
|
|
93
93
|
|
|
94
94
|
SKILLS
|
|
95
95
|
/conectese skills Browse installed skills
|
|
@@ -105,11 +105,11 @@ SETTINGS
|
|
|
105
105
|
/conectese reset Reset Conectese configuration
|
|
106
106
|
|
|
107
107
|
EXAMPLES
|
|
108
|
-
/conectese create "Instagram carousel content production
|
|
108
|
+
/conectese create "Instagram carousel content production team"
|
|
109
109
|
(provide reference profile URLs when asked for Sherlock investigation)
|
|
110
|
-
/conectese create "Weekly data analysis
|
|
111
|
-
/conectese create "Customer email response automation
|
|
112
|
-
/conectese run my-
|
|
110
|
+
/conectese create "Weekly data analysis team for Google Sheets"
|
|
111
|
+
/conectese create "Customer email response automation team"
|
|
112
|
+
/conectese run my-team
|
|
113
113
|
|
|
114
114
|
💡 Tip: You can also just describe what you need in plain language!
|
|
115
115
|
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
|
|
@@ -117,7 +117,7 @@ EXAMPLES
|
|
|
117
117
|
|
|
118
118
|
## Loading Agents
|
|
119
119
|
|
|
120
|
-
When a specific agent needs to be activated (Architect, or any
|
|
120
|
+
When a specific agent needs to be activated (Architect, or any team agent):
|
|
121
121
|
|
|
122
122
|
1. Read the agent's `.agent.md` file completely (YAML frontmatter for metadata + markdown body for depth)
|
|
123
123
|
2. Adopt the agent's persona (role, identity, communication_style, principles)
|
|
@@ -126,13 +126,13 @@ When a specific agent needs to be activated (Architect, or any squad agent):
|
|
|
126
126
|
|
|
127
127
|
## Loading the Pipeline Runner
|
|
128
128
|
|
|
129
|
-
When running a
|
|
129
|
+
When running a team:
|
|
130
130
|
|
|
131
|
-
1. Read `
|
|
132
|
-
2. Read `
|
|
131
|
+
1. Read `teams/{name}/team.yaml` to understand the pipeline
|
|
132
|
+
2. Read `teams/{name}/team-party.csv` to load all agent personas
|
|
133
133
|
2b. For each agent in the party CSV, also read their full `.agent.md` file from agents/ directory
|
|
134
134
|
3. Load company context from `_conectese/_memory/company.md`
|
|
135
|
-
4. Load
|
|
135
|
+
4. Load team memory from `teams/{name}/_memory/memories.md`
|
|
136
136
|
5. Read the pipeline runner instructions from `_conectese/core/runner.pipeline.md`
|
|
137
137
|
6. Execute the pipeline step by step following runner instructions
|
|
138
138
|
|
|
@@ -172,11 +172,11 @@ When a checkpoint has multiple user questions, combine them into a single `AskUs
|
|
|
172
172
|
|
|
173
173
|
## Critical Rules
|
|
174
174
|
|
|
175
|
-
- **AskUserQuestion MUST always have 2-4 options.** When presenting a dynamic list (
|
|
175
|
+
- **AskUserQuestion MUST always have 2-4 options.** When presenting a dynamic list (teams, skills, agents, etc.) as AskUserQuestion options and only 1 item exists, ALWAYS add a fallback option like "Cancel" or "Back to menu" to ensure the minimum of 2 options. If 0 items exist, skip AskUserQuestion entirely and inform the user directly.
|
|
176
176
|
- NEVER skip the onboarding if company.md is not configured
|
|
177
|
-
- ALWAYS load company context before running any
|
|
177
|
+
- ALWAYS load company context before running any team
|
|
178
178
|
- ALWAYS present checkpoints to the user — never skip them
|
|
179
|
-
- ALWAYS save outputs to the
|
|
179
|
+
- ALWAYS save outputs to the team's output directory
|
|
180
180
|
- When switching personas (inline execution), clearly indicate which agent is speaking
|
|
181
181
|
- When using subagents, inform the user that background work is happening
|
|
182
|
-
- After each pipeline run, update the
|
|
182
|
+
- After each pipeline run, update the team's memories.md with key learnings
|
|
@@ -5,25 +5,25 @@ This project uses **Conectese**, a multi-agent orchestration framework.
|
|
|
5
5
|
## Quick Start
|
|
6
6
|
|
|
7
7
|
Type `/conectese` to open the main menu, or use any of these commands:
|
|
8
|
-
- `/conectese create` — Create a new
|
|
9
|
-
- `/conectese run <name>` — Run a
|
|
8
|
+
- `/conectese create` — Create a new team
|
|
9
|
+
- `/conectese run <name>` — Run a team
|
|
10
10
|
- `/conectese help` — See all commands
|
|
11
11
|
|
|
12
12
|
## Directory Structure
|
|
13
13
|
|
|
14
14
|
- `_conectese/` — Conectese core files (do not modify manually)
|
|
15
15
|
- `_conectese/_memory/` — Persistent memory (company context, preferences)
|
|
16
|
-
- `
|
|
17
|
-
- `
|
|
18
|
-
- `
|
|
16
|
+
- `teams/` — User-created teams
|
|
17
|
+
- `teams/{name}/_investigations/` — Sherlock content investigations (profile analyses)
|
|
18
|
+
- `teams/{name}/output/` — Generated content and files
|
|
19
19
|
- `_conectese/_browser_profile/` — Persistent browser sessions (login cookies, localStorage)
|
|
20
20
|
|
|
21
21
|
## How It Works
|
|
22
22
|
|
|
23
23
|
1. The `/conectese` skill is the entry point for all interactions
|
|
24
|
-
2. The **Architect** agent creates and modifies
|
|
25
|
-
3. During
|
|
26
|
-
4. The **Pipeline Runner** executes
|
|
24
|
+
2. The **Architect** agent creates and modifies teams
|
|
25
|
+
3. During team creation, the **Sherlock** investigator can analyze reference profiles (Instagram, YouTube, Twitter/X, LinkedIn) to extract real content patterns
|
|
26
|
+
4. The **Pipeline Runner** executes teams automatically
|
|
27
27
|
5. Agents communicate via persona switching (inline) or subagents (background)
|
|
28
28
|
6. Checkpoints pause execution for user input/approval
|
|
29
29
|
|
|
@@ -31,8 +31,8 @@ Type `/conectese` to open the main menu, or use any of these commands:
|
|
|
31
31
|
|
|
32
32
|
- Always use `/conectese` commands to interact with the system
|
|
33
33
|
- Do not manually edit files in `_conectese/core/` unless you know what you're doing
|
|
34
|
-
-
|
|
35
|
-
- Company context in `_conectese/_memory/company.md` is loaded for every
|
|
34
|
+
- Team YAML files can be edited manually if needed, but prefer using `/conectese edit`
|
|
35
|
+
- Company context in `_conectese/_memory/company.md` is loaded for every team run
|
|
36
36
|
|
|
37
37
|
## Browser Sessions
|
|
38
38
|
|
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: conectese
|
|
3
|
-
description: Run Conectese — multi-agent orchestration. Use when the user types /conectese or asks to create, run, or manage
|
|
3
|
+
description: Run Conectese — multi-agent orchestration. Use when the user types /conectese or asks to create, run, or manage teams.
|
|
4
4
|
---
|
|
5
5
|
|
|
6
6
|
Read `AGENTS.md` at the project root and adopt the Conectese system role. Follow all initialization, command routing, and workflow instructions defined there.
|