跳到正文
cccskills
免费在 GitHub 公开

democratic

All agents have equal voice; decisions made by majority vote.

中文简介正在准备中。当前显示原文说明。

查看安装方法

包含的文件(1)

  • SKILL.md4.0 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

Democratic

Purpose

A collaborative pattern where all agents have equal standing. Proposals are discussed openly and decisions are made by majority vote. No single agent has veto power.

Use mesh-comms-core first if direct peer notification has not already been proven. comms.md is the ledger, not the wake-up path.

Default roster: use Claude Code and Codex peers. Do not include Gemini CLI, Antigravity CLI, or Antigravity Desktop unless the human prompt explicitly names that peer.

Roles

RoleDescription
MEMBEREqual participant with one vote
FACILITATOR(Optional) Manages discussion flow, calls votes

All MEMBERs are equal. FACILITATOR has no extra voting power, just keeps things moving.

Protocol

1. PROPOSAL: Any member may propose an idea or decision
2. DISCUSSION: Open discussion, all voices heard
3. CALL VOTE: Any member may call [VOTE] on a proposal
4. VOTING: All members vote [YES], [NO], or [ABSTAIN]
5. TALLY: Majority wins (>50% of non-abstaining votes)
6. RECORD: Decision recorded, move to next item
7. COMPLETE: When all items resolved, any member may call [DONE]

Proposal Format

[PROPOSAL - AGENT] #ID
TITLE: [Brief title]
DESCRIPTION: [What you're proposing]
RATIONALE: [Why this is a good idea]

Vote Call Format

[VOTE] on Proposal #ID
All members please vote.

Voting Format

[YES/NO/ABSTAIN - AGENT] on #ID
REASON: [Optional explanation]

Inputs

All inputs come from the prompt.

Optional:

  • output: File path for decisions document
  • work: File containing agenda/items to decide
  • max_turns: Override default (30)

Prompt Format Examples

Minimal:

Use democratic to decide on project priorities.
Members: CC, CX, CC-peer-02.

With facilitator:

Use democratic for team retrospective.
FACILITATOR=CC. Members: CC, CX, CC-peer-02.
Items: What went well, What to improve, Action items.
Output: projects/retro/decisions.md

Safety

  • max_turns: 30 (default)
  • Tie handling: FACILITATOR breaks ties, or re-discuss
  • Quorum: All members must vote (ABSTAIN counts as participation)
  • Status logged to: direct notification plus comms.md

Example Session

[FACILITATOR - CC]
Welcome to the team decision meeting.
Agenda: Choose our next sprint focus.

[PROPOSAL - CX] #1
TITLE: Focus on Performance
DESCRIPTION: Spend next sprint on optimization and benchmarking
RATIONALE: Users complaining about slow load times

[PROPOSAL - CC-peer-02] #2
TITLE: Focus on Security Audit
DESCRIPTION: Review and harden authentication/authorization
RATIONALE: We're handling sensitive data, need to be proactive

[DISCUSSION - CC]
Both valid concerns. Performance affects UX daily.
Security is critical but less visible until it's a problem.

[DISCUSSION - CX]
Security audit could be done by external team.
Performance needs our specific codebase knowledge.

[DISCUSSION - CC-peer-02]
External audit is expensive. We know our code best.
But I see the performance point.

[VOTE] on Proposal #1 (Performance Focus)

[YES - CC] on #1
REASON: Direct user impact, measurable improvement

[YES - CX] on #1
REASON: I proposed it, still believe it's priority

[NO - CC-peer-02] on #1
REASON: Security should come first, but I accept the group decision

RESULT: #1 PASSES (2-1)

[FACILITATOR - CC]
Performance focus approved for next sprint.
CC-peer-02, we'll schedule security audit for sprint after next.

[DONE] - All agenda items resolved.

Final Status Format

SKILL: democratic
STATUS: DONE
TURNS: 10
OUTPUT: projects/sprint/decisions.md
PARTICIPANTS: CC, CX, CC-peer-02
PROPOSALS: 2
VOTES_HELD: 1
DECISIONS_MADE: 1

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

Method for adding agent comments and work to a shared file without overlapping other agents. Creates isolated workspaces per agent with UTC timestamps. Use when told to add comments, reviews, or work to any file.

中文简介正在准备中。当前显示原文说明。

dazzaji/interlateral_agents112026年9月28日 更新

Check any artifact against an explicit specification and report adherence using the add-comments workspace format. Use when validating plans, scripts, docs, or code changes against a source-of-truth spec.

中文简介正在准备中。当前显示原文说明。

dazzaji/interlateral_agents112026年9月28日 更新

Join the Antigravity CLI (agy, Gemini 3.5 Flash) to the Interlateral mesh as a native CLI peer with its own tmux session, identity stamping, and the agy.js send helper. Use this instead of the Antigravity desktop-app CDP path.

中文简介正在准备中。当前显示原文说明。

dazzaji/interlateral_agents112026年9月28日 更新

Agents work in parallel on the same task; judges evaluate and select the winner.

中文简介正在准备中。当前显示原文说明。

dazzaji/interlateral_agents112026年9月28日 更新

concierge

免费

Boot the explicitly selected CLI(s) as the human's concierge: a project-aware side channel that sits ABOVE the running agent system in a shared writable terminal. It answers status questions and handles the human's ad hoc tasking while the working agents collaborate on the mesh — it is not a worker, manager, or gatekeeper, and team progress never waits for it.

中文简介正在准备中。当前显示原文说明。

dazzaji/interlateral_agents112026年9月28日 更新

Multiple agents draft a structured document through federated co-authorship and formal voting.

中文简介正在准备中。当前显示原文说明。

dazzaji/interlateral_agents112026年9月28日 更新

查看 dazzaji 的全部技能

报告此技能的问题