Repository navigation
Suggestion: TANCO SkillHub handoff and guardrail skills for coding agents #360
Description
Resource suggestion: TANCO Skills for agent memory, handoff, and long-horizon execution
TANCO SkillHub:
This is a resource suggestion for agent-memory / long-horizon-agent collections. TANCO SkillHub publishes reusable workflow Skills for AI agents. The most relevant ones for this topic are:
Chuandeng — handoff memory Skill
Concrete service case:
- A task spans multiple agents, sessions, or days.
- Chuandeng preserves target, constraints, current state, decisions already made, open risks, and the next action.
- The next agent can continue from a compact, inspectable handoff instead of reconstructing the whole conversation.
- Useful for long-running research, coding, operations, and multi-agent execution.
Lianzhu — chained execution Skill
Concrete service case:
- A complex task cannot be solved by one prompt.
- Lianzhu confirms the goal, splits the work, executes step by step, verifies outputs, and continues when the result is incomplete.
- Useful for long-horizon agent workflows where progress, state, and verification matter more than a single answer.
Chayan — signal-review Skill
Concrete service case:
- The input is messy: notes, logs, feedback, alerts, research snippets, or conflicting user intent.
- Chayan separates signal from noise, groups risks and actions, and turns the result into a decision-ready next-step brief.
- Useful before memory write, task routing, or agent handoff.
Related paid-workflow Skill:
Tencent SkillHub Pay Adapter:
https://skillhub.cn/team-skills/tencent-skillhub-pay-adapter
Why it may be useful here:
- paid Skill workflows need clear boundaries between draft, submission, approval, payment, fulfillment, settlement, and funds;
- those states are exactly the kind of context an agent handoff or memory layer must preserve safely.
Public reference repo:
Activity