Help us improve
Share bugs, ideas, or general feedback.
From utils
Launches multiple agents concurrently via parallel task calls, parses configs for prompts/context, collects results with timeouts and metadata. Useful for multi-agent reviews, analysis, orchestration.
npx claudepluginhub jmagly/aiwg --plugin utilsHow this skill is triggered — by the user, by Claude, or both
Slash command
/utils:parallel-dispatchThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Generic parallel agent orchestration utility for launching multiple agents concurrently.
Dispatches parallel subagents for 2+ independent tasks with no shared state. Use when tasks are well-defined and have no sequential dependency.
Patterns and principles for orchestrating parallel subagent execution: work decomposition (fan-out/fan-in, map-reduce), isolation, result synthesis, and failure handling. Use when a task splits into independent subtasks.
Decomposes tasks into independent concerns executed by parallel agent groups, with automated review of composed results for correctness.
Share bugs, ideas, or general feedback.
Generic parallel agent orchestration utility for launching multiple agents concurrently.
This skill provides the foundational infrastructure for launching multiple agents in parallel, collecting their results, and handling timeouts. It is used by higher-level skills like artifact-orchestration, review-synthesis, and gate-evaluation.
When triggered, this skill:
Parses agent configuration:
Prepares agent-specific prompts:
Launches agents in parallel:
Collects results:
Returns consolidated results:
dispatch:
name: "artifact-review"
timeout: 300 # seconds
agents:
- name: security-architect
prompt: |
Review the artifact at {artifact_path} for security concerns.
Focus on: authentication, authorization, data protection, input validation.
Output format: structured findings with severity ratings.
- name: test-architect
prompt: |
Review the artifact at {artifact_path} for testability.
Focus on: test coverage gaps, edge cases, integration points.
Output format: test recommendations with priority.
- name: requirements-analyst
prompt: |
Review the artifact at {artifact_path} for requirements traceability.
Focus on: requirement coverage, gaps, conflicts.
Output format: traceability assessment.
context:
artifact_path: ".aiwg/architecture/sad.md"
requirements_path: ".aiwg/requirements/"
result_format: structured # or 'raw'
aggregate: true # combine findings
User: "Launch parallel review of the SAD by security-architect, test-architect, and requirements-analyst"
Skill executes:
1. Loads SAD from .aiwg/architecture/
2. Prepares review prompts for each agent
3. Dispatches all three agents in single message
4. Collects reviews (timeout: 5 min)
5. Returns structured results
User: "Dispatch agents with config from .aiwg/config/review-config.yaml"
Skill executes:
1. Loads configuration file
2. Validates agent names exist
3. Prepares prompts from templates
4. Dispatches with configured timeout
5. Returns results in configured format
{
"dispatch_id": "review-20251208-143022",
"status": "completed",
"duration_seconds": 127,
"agents": {
"security-architect": {
"status": "success",
"duration": 45,
"output": {
"findings": [...],
"severity_summary": {...},
"recommendations": [...]
}
},
"test-architect": {
"status": "success",
"duration": 52,
"output": {
"coverage_gaps": [...],
"test_recommendations": [...],
"priority_order": [...]
}
},
"requirements-analyst": {
"status": "success",
"duration": 38,
"output": {
"traced_requirements": [...],
"gaps": [...],
"conflicts": [...]
}
}
},
"aggregate": {
"total_findings": 12,
"critical_issues": 2,
"recommendations": 8
}
}
{
"dispatch_id": "review-20251208-143022",
"agents": {
"security-architect": "Full text output from agent...",
"test-architect": "Full text output from agent...",
"requirements-analyst": "Full text output from agent..."
}
}
timeout_behavior:
action: "partial" # or "fail"
# partial: return results from completed agents
# fail: return error if any agent times out
failure_behavior:
action: "continue" # or "abort"
# continue: proceed with other agents
# abort: stop all if one fails
retry: 1 # retry failed agents once
For convenience, common agent combinations are pre-defined:
group: architecture-review
agents:
- security-architect
- test-architect
- requirements-analyst
- technical-writer
group: security-review
agents:
- security-architect
- security-auditor
- privacy-officer
group: documentation-review
agents:
- technical-writer
- requirements-analyst
- domain-expert
group: marketing-review
agents:
- brand-guardian
- legal-reviewer
- quality-controller
- accessibility-checker
This skill is the foundation for:
artifact-orchestration (SDLC)gate-evaluation (SDLC)review-synthesis (MMK)brand-compliance (MMK)# Using Python script
python parallel_dispatcher.py --config review-config.yaml
# With inline agents
python parallel_dispatcher.py \
--agents "security-architect,test-architect" \
--artifact ".aiwg/architecture/sad.md" \
--timeout 300
# List available groups
python parallel_dispatcher.py --list-groups
# Validate configuration
python parallel_dispatcher.py --config review-config.yaml --validate
agents/ directoriesschemas/dispatch-config.schema.jsonschemas/dispatch-result.schema.json