@orchestrator-claude/definitions 3.11.0 → 3.12.0
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/manifest.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
|
1
1
|
{
|
|
2
2
|
"schema": "orchestrator-definitions/v1",
|
|
3
3
|
"namespace": "orchestrator",
|
|
4
|
-
"version": "3.
|
|
5
|
-
"generatedAt": "2026-03-
|
|
4
|
+
"version": "3.12.0",
|
|
5
|
+
"generatedAt": "2026-03-15T00:48:41.789Z",
|
|
6
6
|
"definitions": {
|
|
7
7
|
"agents": [
|
|
8
8
|
{
|
|
@@ -123,7 +123,7 @@
|
|
|
123
123
|
"slug": "artifact-production",
|
|
124
124
|
"file": "skills/artifact-production/SKILL.md",
|
|
125
125
|
"version": "1.0.0",
|
|
126
|
-
"checksum": "sha256:
|
|
126
|
+
"checksum": "sha256:eafd1bb93cfb094e4f976f17a6c2bd922b651d32120dca44a64e1a8e05081b43",
|
|
127
127
|
"supersedes": null,
|
|
128
128
|
"metadata": {}
|
|
129
129
|
},
|
|
@@ -219,7 +219,7 @@
|
|
|
219
219
|
"slug": "workflow-coordination",
|
|
220
220
|
"file": "skills/workflow-coordination/SKILL.md",
|
|
221
221
|
"version": "1.0.0",
|
|
222
|
-
"checksum": "sha256:
|
|
222
|
+
"checksum": "sha256:209bb65711b497813859e2183e61ee2f9ad1223a1165f31fcee12c45abf1b798",
|
|
223
223
|
"supersedes": null,
|
|
224
224
|
"metadata": {}
|
|
225
225
|
},
|
|
@@ -227,7 +227,7 @@
|
|
|
227
227
|
"slug": "workflow-status",
|
|
228
228
|
"file": "skills/workflow-status/SKILL.md",
|
|
229
229
|
"version": "1.0.0",
|
|
230
|
-
"checksum": "sha256:
|
|
230
|
+
"checksum": "sha256:96c2afc874e1d783435b8fffb654caea5307338d5f53ccfa56ce9d17a097bfbc",
|
|
231
231
|
"supersedes": null,
|
|
232
232
|
"metadata": {}
|
|
233
233
|
}
|
package/package.json
CHANGED
|
@@ -6,10 +6,17 @@ description: How to produce, format, and store workflow artifacts correctly. Use
|
|
|
6
6
|
## Artifact Production Protocol
|
|
7
7
|
|
|
8
8
|
### Storage: MinIO via MCP (NOT filesystem)
|
|
9
|
-
|
|
9
|
+
|
|
10
|
+
Artifacts are stored in MinIO object storage, never on the filesystem.
|
|
11
|
+
The `.orchestrator/artifacts/` directory does NOT exist — all access is via MCP tools.
|
|
12
|
+
|
|
10
13
|
1. Call ToolSearch("select:mcp__orchestrator-extended__artifactStore")
|
|
11
14
|
2. Call artifactStore with: { workflowId, phase, artifactType, content, metadata }
|
|
12
15
|
|
|
16
|
+
To retrieve an artifact:
|
|
17
|
+
1. Call ToolSearch("select:mcp__orchestrator-extended__artifactRetrieve")
|
|
18
|
+
2. Call artifactRetrieve with: { workflowId, phase }
|
|
19
|
+
|
|
13
20
|
### Required Sections by Artifact Type
|
|
14
21
|
|
|
15
22
|
**spec.md (SPECIFY phase):**
|
|
@@ -17,23 +24,44 @@ Artifacts are stored in MinIO, never on filesystem.
|
|
|
17
24
|
- Non-Functional Requirements (performance, security)
|
|
18
25
|
- Architecture Decisions (with rationale)
|
|
19
26
|
- Acceptance Criteria (measurable)
|
|
27
|
+
- Out of Scope (explicit exclusions)
|
|
20
28
|
|
|
21
29
|
**plan.md (PLAN phase):**
|
|
22
|
-
- Architecture Overview (references spec)
|
|
23
|
-
- Component Design (layers, dependencies)
|
|
30
|
+
- Architecture Overview (references spec requirements by number)
|
|
31
|
+
- Component Design (layers, dependencies, interfaces)
|
|
24
32
|
- Implementation Strategy (order, parallelization)
|
|
25
33
|
- Risk Assessment (with mitigations)
|
|
34
|
+
- Testing Strategy (unit, integration, e2e)
|
|
26
35
|
|
|
27
36
|
**tasks.md (TASKS phase):**
|
|
28
|
-
- Atomic tasks (1-3h each)
|
|
29
|
-
- Dependency graph
|
|
30
|
-
- Task groups (parallelizable)
|
|
31
|
-
- Acceptance criteria per task
|
|
37
|
+
- Atomic tasks (1-3h each, single responsibility)
|
|
38
|
+
- Dependency graph (which tasks block which)
|
|
39
|
+
- Task groups (parallelizable sets of 3-5 tasks)
|
|
40
|
+
- Acceptance criteria per task (testable)
|
|
41
|
+
- Checkpoint markers after each task group
|
|
42
|
+
|
|
43
|
+
**implementation-report.md (IMPLEMENT phase):**
|
|
44
|
+
- Tasks completed (with status)
|
|
45
|
+
- Test coverage summary
|
|
46
|
+
- Files changed
|
|
47
|
+
- Checkpoints created
|
|
32
48
|
|
|
33
49
|
### Metadata Template
|
|
34
|
-
|
|
50
|
+
|
|
51
|
+
Always include metadata when storing:
|
|
52
|
+
```json
|
|
53
|
+
{
|
|
54
|
+
"author": "<agent-name>",
|
|
55
|
+
"version": "1.0",
|
|
56
|
+
"status": "draft"
|
|
57
|
+
}
|
|
58
|
+
```
|
|
59
|
+
|
|
60
|
+
The artifact-validator hook will automatically validate after storage.
|
|
61
|
+
Status transitions: draft → validated (after passing validation).
|
|
35
62
|
|
|
36
63
|
### Invocation Tracking
|
|
37
|
-
|
|
64
|
+
|
|
65
|
+
After storing artifact, complete the agent invocation:
|
|
38
66
|
1. ToolSearch("select:mcp__orchestrator-tools__completeAgentInvocation")
|
|
39
67
|
2. completeAgentInvocation with result summary
|
|
@@ -16,19 +16,25 @@ Note: ToolSearch results are cached — you only need to fetch each tool once pe
|
|
|
16
16
|
### Starting a Workflow
|
|
17
17
|
1. Call ToolSearch to load startWorkflow schema
|
|
18
18
|
2. Call startWorkflow with: { name, type, projectId }
|
|
19
|
-
3.
|
|
20
|
-
4. Return
|
|
19
|
+
3. startWorkflow automatically creates the workflow AND sets the first pendingAction
|
|
20
|
+
4. Return the nextStep instruction to the main agent
|
|
21
21
|
|
|
22
|
-
###
|
|
23
|
-
1. Call evaluateGate to check if current phase passes
|
|
24
|
-
2. If PASS: call advancePhase, then setPendingAction for next agent
|
|
25
|
-
3. If FAIL: return blockers list to main agent
|
|
22
|
+
### Deterministic Hooks (ADR-013)
|
|
26
23
|
|
|
27
|
-
|
|
28
|
-
|
|
29
|
-
|
|
30
|
-
|
|
31
|
-
-
|
|
24
|
+
Orchestration governance is enforced by hooks, not by prompt instructions:
|
|
25
|
+
|
|
26
|
+
| Hook | When | Effect |
|
|
27
|
+
|------|------|--------|
|
|
28
|
+
| `ping-pong-enforcer` | After every Agent call | Auto-calls getNextAction, injects result |
|
|
29
|
+
| `gate-guardian` | Before advancePhase | Auto-evaluates gate, blocks if fails |
|
|
30
|
+
| `dangling-workflow-guard` | On session Stop | Warns and completes dangling workflows |
|
|
31
|
+
|
|
32
|
+
You do NOT need to manually call getNextAction or evaluateGate — hooks handle this.
|
|
33
|
+
|
|
34
|
+
### Phase Advancement
|
|
35
|
+
1. Call advancePhase — the gate-guardian hook evaluates the gate automatically
|
|
36
|
+
2. If gate PASSES: advancePhase proceeds, set pendingAction for next agent
|
|
37
|
+
3. If gate FAILS: hook blocks the call and returns failure reasons
|
|
32
38
|
|
|
33
39
|
### Phase → Agent Mapping
|
|
34
40
|
|
|
@@ -41,6 +47,12 @@ Note: ToolSearch results are cached — you only need to fetch each tool once pe
|
|
|
41
47
|
| implement | implementer | implementation-report | markdown |
|
|
42
48
|
| validate | reviewer | validation-report | markdown |
|
|
43
49
|
|
|
50
|
+
### Gate Evaluation Criteria
|
|
51
|
+
- SPECIFY: spec.md exists in MinIO, has required sections
|
|
52
|
+
- PLAN: plan.md exists, references spec, has architecture section
|
|
53
|
+
- TASKS: tasks.md exists, tasks are atomic, has acceptance criteria
|
|
54
|
+
- IMPLEMENT: requires human approval (status: awaiting_approval)
|
|
55
|
+
|
|
44
56
|
### Observability
|
|
45
57
|
After EVERY action, call startAgentInvocation/completeAgentInvocation
|
|
46
58
|
to track your own execution in PostgreSQL.
|
|
@@ -1,322 +1,90 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: workflow-status
|
|
3
3
|
description: Consulta status detalhado do workflow atual incluindo fase, artefatos, gates e proximos passos. Use para verificar progresso ou diagnosticar problemas.
|
|
4
|
-
allowed-tools: Read, Grep, Glob
|
|
5
4
|
---
|
|
6
5
|
|
|
7
6
|
# Workflow Status Skill
|
|
8
7
|
|
|
9
|
-
##
|
|
8
|
+
## Data Source
|
|
10
9
|
|
|
11
|
-
|
|
10
|
+
All workflow state is in **PostgreSQL** via MCP tools. NEVER read `.orchestrator/orchestrator-index.json`.
|
|
12
11
|
|
|
13
|
-
|
|
12
|
+
### MCP Tools for Status Queries
|
|
14
13
|
|
|
15
|
-
|
|
16
|
-
|
|
17
|
-
|
|
18
|
-
|
|
19
|
-
|
|
20
|
-
|
|
21
|
-
|
|
22
|
-
|
|
23
|
-
## Invocacao
|
|
24
|
-
|
|
25
|
-
Exemplos de como invocar esta skill:
|
|
14
|
+
| Tool | What it returns |
|
|
15
|
+
|------|----------------|
|
|
16
|
+
| `getStatus` | Current workflow: phase, status, pending action |
|
|
17
|
+
| `getWorkflows` | All workflows with phase configs |
|
|
18
|
+
| `getNextAction` | Next pending action (agent + status) |
|
|
19
|
+
| `getInvocations` | Agent invocation history |
|
|
20
|
+
| `getMetrics` | Workflow metrics (duration, counts) |
|
|
21
|
+
| `listCheckpoints` | Checkpoint history |
|
|
26
22
|
|
|
23
|
+
Load tools via ToolSearch before first use:
|
|
27
24
|
```
|
|
28
|
-
"
|
|
29
|
-
|
|
30
|
-
"Mostre o progresso da implementacao"
|
|
31
|
-
|
|
32
|
-
"Quais artefatos ja foram gerados?"
|
|
33
|
-
|
|
34
|
-
"Em qual fase estamos?"
|
|
35
|
-
|
|
36
|
-
"Quais sao os proximos passos?"
|
|
37
|
-
|
|
38
|
-
"Liste os checkpoints disponiveis"
|
|
25
|
+
ToolSearch("select:mcp__orchestrator-tools__getStatus")
|
|
39
26
|
```
|
|
40
27
|
|
|
41
|
-
##
|
|
42
|
-
|
|
43
|
-
| Parametro | Tipo | Obrigatorio | Descricao |
|
|
44
|
-
|-----------|------|-------------|-----------|
|
|
45
|
-
| `workflowId` | string | Nao | ID do workflow (usa atual se omitido) |
|
|
46
|
-
| `detail_level` | enum | Nao | summary, standard, detailed |
|
|
47
|
-
| `include_artifacts` | boolean | Nao | Incluir lista de artefatos |
|
|
48
|
-
| `include_checkpoints` | boolean | Nao | Incluir lista de checkpoints |
|
|
49
|
-
| `include_gates` | boolean | Nao | Incluir status dos gates |
|
|
50
|
-
|
|
51
|
-
### Valores de detail_level
|
|
52
|
-
|
|
53
|
-
- **summary**: Apenas fase atual e status geral
|
|
54
|
-
- **standard** (default): Fase, status, ultimos artefatos, proximo passo
|
|
55
|
-
- **detailed**: Tudo incluindo historico completo
|
|
56
|
-
|
|
57
|
-
## Output
|
|
58
|
-
|
|
59
|
-
A skill retorna informacoes estruturadas:
|
|
60
|
-
|
|
61
|
-
```json
|
|
62
|
-
{
|
|
63
|
-
"success": true,
|
|
64
|
-
"workflow": {
|
|
65
|
-
"id": "wf-123",
|
|
66
|
-
"type": "feature_development",
|
|
67
|
-
"feature": "OAuth2 Authentication",
|
|
68
|
-
"status": "in_progress",
|
|
69
|
-
"startedAt": "2025-12-09T10:00:00Z",
|
|
70
|
-
"updatedAt": "2025-12-09T14:30:00Z"
|
|
71
|
-
},
|
|
28
|
+
## When to Use
|
|
72
29
|
|
|
73
|
-
|
|
74
|
-
|
|
75
|
-
|
|
76
|
-
|
|
77
|
-
|
|
78
|
-
"totalTasks": 10,
|
|
79
|
-
"completedTasks": 6,
|
|
80
|
-
"percentage": 60
|
|
81
|
-
}
|
|
82
|
-
},
|
|
30
|
+
- Check current workflow phase and progress
|
|
31
|
+
- Verify which artifacts have been generated
|
|
32
|
+
- Diagnose why a workflow is stuck
|
|
33
|
+
- List available checkpoints
|
|
34
|
+
- Review agent invocation history
|
|
83
35
|
|
|
84
|
-
|
|
85
|
-
{
|
|
86
|
-
"name": "research",
|
|
87
|
-
"status": "completed",
|
|
88
|
-
"duration": "45m",
|
|
89
|
-
"artifacts": ["research-context.md"]
|
|
90
|
-
},
|
|
91
|
-
{
|
|
92
|
-
"name": "specify",
|
|
93
|
-
"status": "completed",
|
|
94
|
-
"duration": "1h",
|
|
95
|
-
"artifacts": ["spec.md"]
|
|
96
|
-
},
|
|
97
|
-
{
|
|
98
|
-
"name": "plan",
|
|
99
|
-
"status": "completed",
|
|
100
|
-
"duration": "1h 30m",
|
|
101
|
-
"artifacts": ["plan.md"]
|
|
102
|
-
},
|
|
103
|
-
{
|
|
104
|
-
"name": "tasks",
|
|
105
|
-
"status": "completed",
|
|
106
|
-
"duration": "45m",
|
|
107
|
-
"artifacts": ["tasks.md"]
|
|
108
|
-
},
|
|
109
|
-
{
|
|
110
|
-
"name": "implement",
|
|
111
|
-
"status": "in_progress",
|
|
112
|
-
"duration": "2h (ongoing)",
|
|
113
|
-
"artifacts": ["impl-001.md", "impl-002.md"]
|
|
114
|
-
}
|
|
115
|
-
],
|
|
36
|
+
## How to Report Status
|
|
116
37
|
|
|
117
|
-
|
|
118
|
-
{
|
|
119
|
-
"id": "art-001",
|
|
120
|
-
"type": "research",
|
|
121
|
-
"path": ".orchestrator/artifacts/research/research-context.md",
|
|
122
|
-
"status": "validated",
|
|
123
|
-
"createdAt": "2025-12-09T10:45:00Z"
|
|
124
|
-
},
|
|
125
|
-
{
|
|
126
|
-
"id": "art-002",
|
|
127
|
-
"type": "specification",
|
|
128
|
-
"path": ".orchestrator/artifacts/specify/spec.md",
|
|
129
|
-
"status": "validated",
|
|
130
|
-
"createdAt": "2025-12-09T11:45:00Z"
|
|
131
|
-
}
|
|
132
|
-
],
|
|
133
|
-
|
|
134
|
-
"gates": {
|
|
135
|
-
"research": { "passed": true, "evaluatedAt": "2025-12-09T10:50:00Z" },
|
|
136
|
-
"specify": { "passed": true, "evaluatedAt": "2025-12-09T11:50:00Z" },
|
|
137
|
-
"plan": { "passed": true, "evaluatedAt": "2025-12-09T13:20:00Z" },
|
|
138
|
-
"tasks": { "passed": true, "evaluatedAt": "2025-12-09T14:00:00Z" },
|
|
139
|
-
"implement": { "passed": false, "pending": true }
|
|
140
|
-
},
|
|
141
|
-
|
|
142
|
-
"checkpoints": [
|
|
143
|
-
{
|
|
144
|
-
"id": "chk-001",
|
|
145
|
-
"phase": "specify",
|
|
146
|
-
"message": "[orchestrator] specify: specification validated",
|
|
147
|
-
"gitRef": "abc123",
|
|
148
|
-
"createdAt": "2025-12-09T11:50:00Z"
|
|
149
|
-
},
|
|
150
|
-
{
|
|
151
|
-
"id": "chk-002",
|
|
152
|
-
"phase": "plan",
|
|
153
|
-
"message": "[orchestrator] plan: technical plan approved",
|
|
154
|
-
"gitRef": "def456",
|
|
155
|
-
"createdAt": "2025-12-09T13:20:00Z"
|
|
156
|
-
}
|
|
157
|
-
],
|
|
158
|
-
|
|
159
|
-
"nextSteps": [
|
|
160
|
-
"Completar TASK-007: Implementar middleware de autenticacao",
|
|
161
|
-
"Executar testes de integracao",
|
|
162
|
-
"Avaliar gate de implement para concluir workflow"
|
|
163
|
-
],
|
|
164
|
-
|
|
165
|
-
"metrics": {
|
|
166
|
-
"totalDuration": "4h 30m",
|
|
167
|
-
"estimatedRemaining": "2h",
|
|
168
|
-
"artifactsGenerated": 6,
|
|
169
|
-
"gatesPassed": 4,
|
|
170
|
-
"checkpointsCreated": 4
|
|
171
|
-
}
|
|
172
|
-
}
|
|
173
|
-
```
|
|
174
|
-
|
|
175
|
-
## Formato de Output para Usuario
|
|
176
|
-
|
|
177
|
-
Quando invocada, a skill formata output amigavel:
|
|
38
|
+
### Quick Status (default)
|
|
178
39
|
|
|
179
40
|
```markdown
|
|
180
|
-
## Status
|
|
181
|
-
|
|
182
|
-
**ID**: wf-123
|
|
183
|
-
**Tipo**: feature_development
|
|
184
|
-
**Feature**: OAuth2 Authentication
|
|
185
|
-
**Status**: 🟡 Em Progresso
|
|
186
|
-
|
|
187
|
-
### Fase Atual: IMPLEMENT
|
|
188
|
-
|
|
189
|
-
```
|
|
190
|
-
[████████░░░░░░░░░░░░] 60% (6/10 tasks)
|
|
191
|
-
```
|
|
41
|
+
## Workflow Status
|
|
192
42
|
|
|
193
|
-
|
|
43
|
+
**ID**: wf-123 | **Type**: feature_development | **Status**: in_progress
|
|
44
|
+
**Phase**: IMPLEMENT (6/10 tasks)
|
|
194
45
|
|
|
195
|
-
|
|
196
|
-
|
|
197
|
-
|
|
198
|
-
| specify |
|
|
199
|
-
| plan |
|
|
200
|
-
| tasks |
|
|
201
|
-
| implement |
|
|
46
|
+
### Phase Progress
|
|
47
|
+
| Phase | Status | Gate |
|
|
48
|
+
|-------|--------|------|
|
|
49
|
+
| specify | completed | passed |
|
|
50
|
+
| plan | completed | passed |
|
|
51
|
+
| tasks | completed | passed |
|
|
52
|
+
| implement | in_progress | pending |
|
|
202
53
|
|
|
203
|
-
###
|
|
204
|
-
|
|
205
|
-
- [x] research-complete
|
|
206
|
-
- [x] specification-complete
|
|
207
|
-
- [x] plan-complete
|
|
208
|
-
- [x] tasks-complete
|
|
209
|
-
- [ ] implementation-complete (pendente)
|
|
210
|
-
|
|
211
|
-
### Proximos Passos
|
|
212
|
-
|
|
213
|
-
1. Completar TASK-007: Implementar middleware de autenticacao
|
|
214
|
-
2. Executar testes de integracao
|
|
215
|
-
3. Avaliar gate de implement para concluir workflow
|
|
216
|
-
|
|
217
|
-
### Metricas
|
|
218
|
-
|
|
219
|
-
- **Tempo Total**: 4h 30m
|
|
220
|
-
- **Estimativa Restante**: 2h
|
|
221
|
-
- **Artefatos**: 6
|
|
222
|
-
- **Checkpoints**: 4
|
|
54
|
+
### Next Action
|
|
55
|
+
Agent: implementer | Status: awaiting_agent
|
|
223
56
|
```
|
|
224
57
|
|
|
225
|
-
|
|
226
|
-
|
|
227
|
-
Esta skill invoca o handler TypeScript:
|
|
58
|
+
### Detailed Status (when user asks for details)
|
|
228
59
|
|
|
229
|
-
|
|
230
|
-
|
|
60
|
+
Add to quick status:
|
|
61
|
+
- **Artifacts**: List with type, status (draft/validated), storage location
|
|
62
|
+
- **Checkpoints**: List with phase, git ref, timestamp
|
|
63
|
+
- **Invocations**: Agent history with duration and result
|
|
64
|
+
- **Metrics**: Total duration, artifacts generated, gates passed
|
|
231
65
|
|
|
232
|
-
|
|
233
|
-
name = 'workflow-status';
|
|
234
|
-
description = 'Consulta status detalhado do workflow';
|
|
66
|
+
## Diagnosing Common Problems
|
|
235
67
|
|
|
236
|
-
|
|
237
|
-
|
|
238
|
-
|
|
239
|
-
|
|
240
|
-
|
|
241
|
-
// 5. Lista artefatos e checkpoints
|
|
242
|
-
// 6. Calcula metricas
|
|
243
|
-
// 7. Formata output
|
|
244
|
-
}
|
|
245
|
-
}
|
|
246
|
-
```
|
|
68
|
+
### Workflow stuck — no pending action
|
|
69
|
+
1. Call `getStatus` to check current phase
|
|
70
|
+
2. Call `getNextAction` — if `hasAction: false`:
|
|
71
|
+
- In implement phase → call `completeWorkflow`
|
|
72
|
+
- In other phases → orchestrator needs to set next pendingAction
|
|
247
73
|
|
|
248
|
-
|
|
249
|
-
|
|
250
|
-
|
|
251
|
-
|
|
252
|
-
|
|
253
|
-
Skill: workflow-status com detail_level=summary
|
|
254
|
-
Output: Fase atual, status, progresso percentual
|
|
255
|
-
```
|
|
256
|
-
|
|
257
|
-
### Relatorio Detalhado
|
|
258
|
-
```
|
|
259
|
-
User: "Me de um relatorio completo do progresso"
|
|
260
|
-
Skill: workflow-status com detail_level=detailed, todos includes=true
|
|
261
|
-
Output: Relatorio completo com historico
|
|
262
|
-
```
|
|
74
|
+
### Gate not passing
|
|
75
|
+
1. Call `getStatus` to see which phase is current
|
|
76
|
+
2. Check artifact exists for that phase via `artifactRetrieve`
|
|
77
|
+
3. Check artifact status (draft vs validated)
|
|
78
|
+
4. If draft → artifact-validator hook may not have fired
|
|
263
79
|
|
|
264
|
-
###
|
|
265
|
-
|
|
266
|
-
|
|
267
|
-
|
|
268
|
-
Output: Lista de artefatos com status e paths
|
|
269
|
-
```
|
|
270
|
-
|
|
271
|
-
### Diagnostico de Problema
|
|
272
|
-
```
|
|
273
|
-
User: "Por que o workflow esta travado?"
|
|
274
|
-
Skill: workflow-status com include_gates=true
|
|
275
|
-
Output: Status detalhado mostrando gate pendente
|
|
276
|
-
```
|
|
277
|
-
|
|
278
|
-
## Comportamento em Casos Especiais
|
|
279
|
-
|
|
280
|
-
### Nenhum Workflow Ativo
|
|
281
|
-
```json
|
|
282
|
-
{
|
|
283
|
-
"success": false,
|
|
284
|
-
"error": {
|
|
285
|
-
"code": "NO_ACTIVE_WORKFLOW",
|
|
286
|
-
"message": "Nenhum workflow ativo encontrado",
|
|
287
|
-
"suggestion": "Use 'orchestrate start' para iniciar um novo workflow"
|
|
288
|
-
}
|
|
289
|
-
}
|
|
290
|
-
```
|
|
291
|
-
|
|
292
|
-
### Workflow Nao Encontrado
|
|
293
|
-
```json
|
|
294
|
-
{
|
|
295
|
-
"success": false,
|
|
296
|
-
"error": {
|
|
297
|
-
"code": "WORKFLOW_NOT_FOUND",
|
|
298
|
-
"message": "Workflow 'wf-xyz' nao encontrado",
|
|
299
|
-
"suggestion": "Verifique o ID ou use sem parametro para workflow atual"
|
|
300
|
-
}
|
|
301
|
-
}
|
|
302
|
-
```
|
|
303
|
-
|
|
304
|
-
### Index Corrompido
|
|
305
|
-
```json
|
|
306
|
-
{
|
|
307
|
-
"success": false,
|
|
308
|
-
"error": {
|
|
309
|
-
"code": "INDEX_CORRUPTED",
|
|
310
|
-
"message": "orchestrator-index.json esta corrompido ou invalido",
|
|
311
|
-
"suggestion": "Execute 'orchestrator config validate' para diagnosticar"
|
|
312
|
-
}
|
|
313
|
-
}
|
|
314
|
-
```
|
|
80
|
+
### Phase stuck in in_progress
|
|
81
|
+
1. Check if agent invocation completed (`getInvocations`)
|
|
82
|
+
2. Check if `completeAgentInvocation` was called
|
|
83
|
+
3. If not → pending action may be stale
|
|
315
84
|
|
|
316
|
-
##
|
|
85
|
+
## Important Notes
|
|
317
86
|
|
|
318
|
-
1.
|
|
319
|
-
2.
|
|
320
|
-
3.
|
|
321
|
-
4.
|
|
322
|
-
5. Cache de 30 segundos para evitar leituras repetidas
|
|
87
|
+
1. This skill is **read-only** — it queries state, never modifies it
|
|
88
|
+
2. All data comes from PostgreSQL via MCP tools
|
|
89
|
+
3. Artifacts content is in MinIO — use `artifactRetrieve` to read
|
|
90
|
+
4. The `ping-pong-enforcer` hook auto-injects getNextAction after Agent calls
|