kanban-dispatcher
SQLite task board for structured multi-agent parallelization. Workers claim tasks, send heartbeats, request approval gates, and spawn child tasks. Use when orchestrating 3+ parallel agents on a complex project with dependencies.
- License 许可
- MIT
- 最近更新
- 2026-11-14
- Stars
- ★ 13
- Forks
- 2
01 前言 Front Matter
- name
- kanban-dispatcher
- source
- https://github.com/NousResearch/hermes-agent
- license
- MIT
- description
- SQLite task board for structured multi-agent parallelization. Workers claim tasks, send heartbeats, request approval gates, and spawn child tasks. Use when orchestrating 3+ parallel agents on a complex project with dependencies.
03 SKILL.md 正文
Kanban Dispatcher
Structured task board that prevents parallel agents from stepping on each other. Each worker claims one task, heartbeats to stay alive, and can spawn child tasks.
Trigger phrases: "parallel tasks", "kanban", "task board", "orchestrate agents", "dispatch workers", "multi-agent project"
Core State Machine
PENDING → IN_PROGRESS (worker claims)
→ BLOCKED (needs human input)
→ DONE (worker completes)
→ FAILED (worker errors)
Worker Contract
Every worker agent MUST follow this sequence:
# 1. Show current task at session start
kanban_show()
# 2. Heartbeat every 2-3 tool calls (proves liveness)
kanban_heartbeat(note="Running tests in src/auth/")
# 3. If blocked, surface to human
kanban_block(reason="Missing API key — need STRIPE_SECRET_KEY in .env")
# 4. Spawn child task if work decomposes
kanban_create(
title="Fix type errors in auth module",
assignee="implementer",
parents=["TASK-001"], # blocks parent until done
context="See src/auth/types.ts line 42"
)
# 5. Complete with structured output
kanban_complete(
summary="Implemented OAuth2 flow, 14 tests pass",
metadata={"files_changed": ["src/auth/oauth.ts"], "tests_added": 14}
)
Orchestrator Pattern
# Create task board
kanban_create(title="Implement user auth", assignee="implementer")
kanban_create(title="Write auth tests", assignee="tester", parents=["TASK-001"])
kanban_create(title="Auth code review", assignee="reviewer", parents=["TASK-002"])
# Dispatch workers in parallel
delegate_task(goal="Work the kanban board as implementer", role="implementer")
delegate_task(goal="Work the kanban board as tester", role="tester")
# Workers auto-sequence via parent dependencies
# tester waits for implementer to complete TASK-001
# reviewer waits for tester to complete TASK-002
Approval Gate
kanban_request_approval(
task_id="TASK-003",
summary="About to apply irreversible migration — please confirm",
context="Backup at /tmp/backup.sql"
)
# Worker blocks here until human calls kanban_approve("TASK-003")
Heartbeat Protocol
Workers MUST heartbeat — stale tasks get reclaimed after 5 minutes:
# Call every 2-3 tool calls during long operations
kanban_heartbeat(note="Still running cargo build...")
Task Roles
| Role | Toolsets | Focus |
|---|---|---|
implementer |
terminal, file | Write code, run tests |
tester |
terminal, file | Write tests, verify coverage |
reviewer |
read-only | Spec compliance + code quality |
documenter |
file | Update docs, changelogs |
Anti-Fake-Pass
❌ Worker never heartbeats — task reclaimed after timeout
❌ Completing task before child tasks finish
❌ > 3 levels of child task nesting
❌ Blocking without a human-resolvable question
❌ Summary "done" — not a summary
See Also
core/skills/claude-swarm-orchestration/SKILL.md— auto-decompose approachcore/skills/subagent-dependency/SKILL.md— DAG-based orchestration
04 同仓库更多 Skills
openai--sentry--sentry
Use when the user asks to inspect Sentry issues or events, summarize recent production errors, or pull basic Sentry health data via the Sentry API; perform read-only queries with the bundled script and require `SENTRY_AUTH_TOKEN`.
ito-basket-compare
Compare Itô prediction-market baskets against a user's knowledge base, portfolio notes, financial context, watchlist, or research thesis. Use for read-only basket comparison and gap analysis without investment advice or live trading.
implementing-network-traffic-baselining
Build network traffic baselines from NetFlow/IPFIX data using Python pandas for statistical analysis, z-score anomaly detection, and hourly/daily traffic pattern profiling
detecting-evasion-techniques-in-endpoint-logs
Detects defense evasion techniques used by adversaries in endpoint logs including log tampering, timestomping, process injection, and security tool disabling. Use when investigating suspicious endpoint behavior, building detection rules for evasion tactics, or conducting threat hunting for stealthy adversary activity. Activates for requests involving evasion detection, defense evasion analysis, log tampering detection, or MITRE ATT&CK TA0005.
implementing-microsegmentation-with-guardicore
Implementing microsegmentation using Akamai Guardicore Segmentation to map application dependencies, create granular network policies, visualize east-west traffic flows, and enforce least-privilege communication between workloads across data centers and cloud.
ramdisk-tmpfs-patterns
暂无简介