AGENT SKILL

add-member

Add a new team member. Describe what you need and ATeam infers the config.

License 许可
Apache-2.0
最近更新
2026-11-14
Stars
★ 1.6k
Forks
491

01 前言 Front Matter

name
add-member
description
Add a new team member. Describe what you need and ATeam infers the config.

03 SKILL.md 正文

AgenTeam Add Role

Add a new team member from a natural language description.

Process

1. Auto-Init Guard

Check for .agenteam/config.yaml, .agenteam.team/config.yaml, or legacy agenteam.yaml. If all are missing:

  • Create config dir: mkdir -p .agenteam
  • Copy the template: cp <plugin-dir>/templates/agenteam.yaml.template .agenteam/config.yaml
  • Set the team name to the project directory name
  • Generate agents: python3 <runtime>/agenteam_rt.py generate

2. Parse Intent

Extract role details from the user's natural language request. Examples:

  • "Add a performance tuning engineer" -> name: performance_engineer, focus: profiling + optimization
  • "I need a security auditor on the team" -> name: security_auditor, focus: vulnerabilities + auth
  • "Add a docs writer that maintains README and API docs" -> name: docs_writer, focus: documentation, write_scope: docs/**
  • "Add a DevOps engineer for CI/CD" -> name: devops_engineer, focus: pipelines + deployment

Infer as much as possible from the description:

Field How to infer
name Snake_case from the role title
description From user's description
responsibilities 3-5 items inferred from the role's domain
participates_in Match to pipeline stages: research, strategy, design, plan, implement, test, review
can_write Yes if the role creates/modifies files; no if it only analyzes
write_scope Infer from what the role writes (docs, src, tests, configs)
model Omit to inherit platform default. If the user requests a pin, inspect codex debug models and choose only from the live catalog. Use evaluation evidence to justify a strong model for demanding analysis or a smaller model for bounded worker tasks. Model is a personal override — share-config strips it.
reasoning_effort high for analysis roles, medium for execution/writing roles
system_instructions Generate focused instructions from the role's domain

3. Team Size Check

Before confirming, count the current roles:

python3 <runtime>/agenteam_rt.py roles list
  • 7-12 roles: No warning. This is the productive range.
  • 13+ roles: Warn the user: "Your team will have [N] roles. Teams above 12 can increase coordination overhead and make role selection harder. Consider extending an existing role's system_instructions instead. Proceed anyway? (yes / cancel)"

Also check: if the team will have more than 6 roles, note: "Codex defaults to 6 concurrent agent threads. To run more agents in parallel, set agents.max_threads in your Codex config.toml."

4. Confirm with User

Present the inferred role as a summary and ask for confirmation:

Here's your new team member:

  Name: performance_engineer
  Focus: Profiling, bottleneck analysis, optimization
  Stages: review, implement
  Writes to: src/** (optimization patches)
  Model: inherited default (or `gpt-5.4` if you want to pin it)

  System instructions:
  You are the performance engineer on an AgenTeam. Your primary job is
  to identify bottlenecks and optimize critical paths...

Add to team? (yes / adjust)

If the user says "adjust" or requests changes, update the fields and re-confirm. Do not ask field-by-field -- keep it conversational.

5. Write to Config

Read the current .agenteam/config.yaml (or legacy agenteam.yaml) and add the new role under roles:.

Write the full role block including:

  • description
  • responsibilities
  • participates_in
  • can_write and write_scope (if applicable)
  • model and reasoning_effort
  • parallel_safe (true for read-only roles, false for writers unless scoped)
  • system_instructions

If the role participates in a pipeline stage, also add it to the appropriate pipeline.stages[].roles list.

6. Regenerate Agents

python3 <runtime>/agenteam_rt.py generate

7. Confirm

Show the user:

  • The generated agent file: .codex/agents/<name>.toml
  • How to use it immediately:
    • Codex App: @ateam ask <name> to <task>
    • Codex CLI: $ateam:assign <name> "<task>"
  • Reminder: "Edit .agenteam/config.yaml anytime to adjust this role."

04 同仓库更多 Skills

hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-review-with-annotations

Create, inspect, reply to, update, or resolve situated Worktable annotations. Use when feedback belongs to a Doc section, block, text range, or HTML Doc, or when the user asks to address existing review feedback and coordinate a requested artifact edit.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-find-and-synthesize-context

Find, assess, and synthesize context from Worktable Docs and Records. Use when the user asks what a workspace contains, wants workspace-backed research, needs current facts reconciled across artifacts, or requests a cited synthesis without changing workspace state.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-create-or-update-docs

Create, edit, rename, or maintain narrative Worktable Docs. Use for plans, decisions, briefs, notes, research syntheses, and other durable prose when the user requests a new artifact or a safe update to an existing one.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-create-or-manage-records

Organize repeated structured information as durable Worktable Records. Use for research datasets, inventories, directories, pipelines, trackers, and other collections whose items need identity, fields, filtering, links, validation, or individual updates.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-create-or-manage-interactive-html

Create or improve durable visual and interactive artifacts in Worktable. Use for dashboards, trackers, calculators, mini apps, slide presentations, and other requests that benefit from a polished HTML Doc.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-collaborate-threads

Read, answer, and hand off messages through durable Worktable threads with users or other authenticated participants. Use when the user asks to check addressed messages, reply to a participant, continue a thread, request another agent's help, wait briefly for a response, or watch for comments and messages sent to the agent.

★ 1.6k