TypingMindTypingMindThe best frontend for AI
Agencycli Usage logo

Agencycli Usage

Community
chenhg5
agencycli-usage

Operate the agencycli tool — create agencies, teams, projects, and hire AI agents into working directories.

Overview

Publisherchenhg5
Repositoryagencycli
Skill nameagencycli-usage
Stars
240
Forks
26
Bundled files
Instructions only
Links
Skill sourceRepository
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by chenhg5 on GitHub. Read the source before you install it.

Installation

Install the Agencycli Usage AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/chenhg5/agencycli/tree/main/skills/agencycli-usage
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/chenhg5/agencycli.git /tmp/agencycli
mkdir -p .claude/skills
cp -r /tmp/agencycli/skills/agencycli-usage .claude/skills/agencycli-usage
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Agencycli Usage in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Agencycli Usage on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Agencycli Usage is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Skill: agencycli Usage

agencycli is the CLI tool that builds and manages this agency. Use it to manage teams, projects, agents, tasks, and inter-agent communication.

Workspace

The current agency workspace is at: $AGENCY_DIR

All commands auto-discover the workspace when run from inside it, or use --dir:

bash
agencycli --dir $AGENCY_DIR <command>

Discover what exists

bash
agencycli --dir $AGENCY_DIR list teams      # all teams
agencycli --dir $AGENCY_DIR list projects   # all projects
agencycli --dir $AGENCY_DIR list agents     # all agents across all projects
agencycli --dir $AGENCY_DIR list skills     # available skills

agencycli --dir $AGENCY_DIR show team engineering
agencycli --dir $AGENCY_DIR show project cc-connect
agencycli --dir $AGENCY_DIR show agent cc-connect pm
agencycli --dir $AGENCY_DIR show agent cc-connect pm --raw  # full merged context

Tasks

Add a task for an agent

bash
agencycli --dir $AGENCY_DIR task add \
  --project <project> --agent <agent> \
  --title "Task title" \
  --prompt "Detailed instructions..." \
  --priority <0-3>   # 0=critical 1=high 2=normal(default) 3=low

View the task queue (sorted by priority)

bash
agencycli --dir $AGENCY_DIR task list --project <project> --agent <agent>
agencycli --dir $AGENCY_DIR task list --project <project> --agent <agent> --status pending

Control tasks

bash
agencycli --dir $AGENCY_DIR task set <task-id> [--title T] [--status S] [--priority N] \
  [--due-date YYYY-MM-DD] [--estimate-duration 30m] [--parent ID] [--label L]...
agencycli --dir $AGENCY_DIR task stats [--since today] [--project P] [--agent A] [--assignee X] \
  [--label L] [--by agent|assignee|label|label:value|label:category] [--detail] [--format json]
agencycli --dir $AGENCY_DIR task cancel <task-id>
agencycli --dir $AGENCY_DIR task retry  <task-id>

# Emergency halt — cancel all pending (and optionally running) tasks
agencycli --dir $AGENCY_DIR task stop-all \
  --project <project> --all-agents
agencycli --dir $AGENCY_DIR task stop-all \
  --project <project> --agent <agent> --include-running

View token usage and cost

bash
# One agent
agencycli --dir $AGENCY_DIR task tokens \
  --project <project> --agent <agent>

# All agents in a project
agencycli --dir $AGENCY_DIR task tokens \
  --project <project> --all-agents

# Specific task
agencycli --dir $AGENCY_DIR task tokens \
  --project <project> --agent <agent> --task <task-id>

Task done / confirm-request (called by agents inside their prompt)

bash
# Report completion
agencycli --dir $AGENCY_DIR task done \
  --id $TASK_ID --status success --summary "Brief description of what was done"

# Report failure
agencycli --dir $AGENCY_DIR task done \
  --id $TASK_ID --status failed --error "reason"

# Request human input before proceeding (non-blocking)
agencycli --dir $AGENCY_DIR task confirm-request \
  --id $TASK_ID \
  --summary "PR #42 ready, awaiting merge approval" \
  --action-item "Review: gh pr view 42 --repo org/repo" \
  --action-item "Approve: inbox reply <msg-id> --body 'approve'" \
  --action-item "Hold: inbox reply <msg-id> --body 'hold <reason>'"

After confirm-request, the task is archived (non-blocking). The human replies via inbox reply and the agent continues on their next wakeup.

Inbox — async messages (non-blocking)

Any participant (human or agent) can send non-blocking messages to any other participant. The recipient reads them on their next wakeup — the scheduler automatically injects unread messages at the top of the wakeup prompt.

Participant address format

  • Human: human
  • Agent: <project>/<agent> — e.g. cc-connect/pm, cc-connect/dev-claude

Send a message

bash
# Human → agent
agencycli --dir $AGENCY_DIR inbox send \
  --to <project>/<agent> \
  --subject "Subject" \
  --body "Body"

# Agent → human
agencycli --dir $AGENCY_DIR inbox send \
  --from <project>/<agent> --to human \
  --subject "Subject" --body "Body"

# Agent → agent
agencycli --dir $AGENCY_DIR inbox send \
  --from <project>/pm --to <project>/dev \
  --subject "Extra context for task <id>" \
  --body "Details..."

# Group send (repeat --to for multiple recipients)
agencycli --dir $AGENCY_DIR inbox send \
  --from <project>/pm \
  --to <project>/dev --to <project>/qa --to human \
  --subject "Sprint kick-off" --body "..."

Read messages

bash
agencycli --dir $AGENCY_DIR inbox messages                          # human's unread
agencycli --dir $AGENCY_DIR inbox messages --recipient <project>/pm # agent's mailbox
agencycli --dir $AGENCY_DIR inbox messages --from <project>/qa      # filter by sender
agencycli --dir $AGENCY_DIR inbox messages --all                    # include read
agencycli --dir $AGENCY_DIR inbox messages --archived               # archived only
agencycli --dir $AGENCY_DIR inbox messages --mark-read              # mark all read

Reply / Forward

bash
agencycli --dir $AGENCY_DIR inbox reply <msg-id> --from <address> --body "..."
agencycli --dir $AGENCY_DIR inbox fwd   <msg-id> --to <address> --note "..."

Per-message status

bash
agencycli --dir $AGENCY_DIR inbox read    <msg-id>
agencycli --dir $AGENCY_DIR inbox archive <msg-id>
agencycli --dir $AGENCY_DIR inbox delete  <msg-id>
agencycli --dir $AGENCY_DIR inbox rm      <msg-id>   # alias for delete

Heartbeat scheduler

bash
agencycli --dir $AGENCY_DIR scheduler start
agencycli --dir $AGENCY_DIR scheduler stop
agencycli --dir $AGENCY_DIR scheduler status

# Configure an agent's heartbeat
agencycli --dir $AGENCY_DIR scheduler heartbeat \
  --project <project> --agent <agent> \
  --enable --interval 30m \
  --active-hours "09:00-20:00" \
  --active-days weekdays

# Set or update the wakeup routine (runs when task queue is empty)
agencycli --dir $AGENCY_DIR scheduler heartbeat \
  --project <project> --agent <agent> \
  --wakeup-prompt-file projects/<project>/agents/<agent>/wakeup.md

When the scheduler wakes an agent:

  1. If pending tasks exist → runs the highest-priority task
  2. If queue is empty and a wakeup routine is set → runs wakeup.md as a synthetic task
  3. Any unread messages for the agent are auto-prepended to the prompt

Cron jobs

bash
agencycli --dir $AGENCY_DIR cron add \
  --project <project> --agent <agent> \
  --title "Weekly backlog review" \
  --schedule "0 9 * * 1" \
  --prompt "Review the backlog and reprioritise for the week..."

agencycli --dir $AGENCY_DIR cron list    --project <project> --agent <agent>
agencycli --dir $AGENCY_DIR cron disable <cron-id> --project <project> --agent <agent>

OKRs (Objectives and Key Results)

OKRs support three scope levels: agency (global), project, and agent. Child OKRs can link to a parent via --parent for hierarchical alignment.

bash
# List — filter by scope and/or quarter
agencycli --dir $AGENCY_DIR okr list
agencycli --dir $AGENCY_DIR okr list --quarter 2026-Q2
agencycli --dir $AGENCY_DIR okr list --scope project --scope-ref my-service
agencycli --dir $AGENCY_DIR okr list --scope agent --scope-ref my-service/dev

# Create — specify scope level
agencycli --dir $AGENCY_DIR okr create \
  --objective "Ship v2.0" --owner human --quarter 2026-Q2
agencycli --dir $AGENCY_DIR okr create \
  --objective "Reduce latency" --scope project --scope-ref my-service
agencycli --dir $AGENCY_DIR okr create \
  --objective "Improve autonomy" --scope agent --scope-ref my-service/dev \
  --parent <parent-okr-id> --description "Detailed description"

# Show / Update / Delete
agencycli --dir $AGENCY_DIR okr show <okr-id>
agencycli --dir $AGENCY_DIR okr update <okr-id> --status at_risk
agencycli --dir $AGENCY_DIR okr update <okr-id> --scope project --scope-ref my-service
agencycli --dir $AGENCY_DIR okr delete <okr-id>

# Key Results
agencycli --dir $AGENCY_DIR okr kr add \
  --okr <okr-id> --description "Reduce p95 latency" --target 200 --unit ms
agencycli --dir $AGENCY_DIR okr kr update \
  --okr <okr-id> --kr <kr-id> --current 150

# Review notes
agencycli --dir $AGENCY_DIR okr review \
  --okr <okr-id> --note "Good progress this week" --author human

OKR scope: agency (default), project, agent. Status: on_track, in_progress, at_risk, off_track, achieved. Progress is auto-calculated from Key Results.

When completing a task, check if it advances any Key Result and update accordingly.

Milestones

bash
agencycli --dir $AGENCY_DIR milestone list --project <project>
agencycli --dir $AGENCY_DIR milestone create --project <project> \
  --title "Beta Release" --due-date 2026-05-01 --owner human \
  --criteria "All P0 bugs fixed" --criteria "Docs updated"
agencycli --dir $AGENCY_DIR milestone show <ms-id> --project <project>
agencycli --dir $AGENCY_DIR milestone update <ms-id> --project <project> \
  --progress 75 --status in_progress
agencycli --dir $AGENCY_DIR milestone delete <ms-id> --project <project>

Milestone status: planned, in_progress, completed, cancelled.

Context sync

bash
agencycli --dir $AGENCY_DIR sync                                # all agents with changed context
agencycli --dir $AGENCY_DIR sync --project <project>           # one project
agencycli --dir $AGENCY_DIR sync --project <project> --name <agent>  # one agent
agencycli --dir $AGENCY_DIR sync --force                        # force regenerate everything

Context Inheritance

Every hired agent automatically receives context in this order:

  1. agency-prompt.md — global rules, values
  2. teams/<team>/prompt.md — team-specific standards
  3. teams/<team>/roles/<role>/prompt.md — role responsibilities
  4. projects/<project>/prompt.md — project background and tech stack
  5. skills — all skills listed in the team's team.yaml and role's role.yaml

Frequently asked questions

What does the Agencycli Usage AI skill do?

Operate the agencycli tool — create agencies, teams, projects, and hire AI agents into working directories.

Why use Agencycli Usage on TypingMind?

Because you install it once and use it with any model. Agencycli Usage is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Agencycli Usage in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/chenhg5/agencycli/tree/main/skills/agencycli-usage. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Agencycli Usage?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Agencycli Usage?

As many as you like. As long as a model supports skills, you can use Agencycli Usage with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Agencycli Usage AI skill free?

It is published on GitHub by chenhg5. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

Related AI Skills

View all
HKUDS logo

Cli Anything 3mf

HKUDS
OrganizationPopular

3MF mesh geometry editor — detect and resize cylindrical holes, repair meshes, compare 3D printing files. Works with BambuStudio and PrusaSlicer 3MF files.

4.6K
49.5K
Instructions
View
HKUDS logo

Cli Anything Godot

HKUDS
OrganizationPopular

Agent-native CLI for Godot project management, scenes, exports, and script execution.

4.6K
49.5K
Instructions
View
HKUDS logo

Cli Anything Intelwatch

HKUDS
OrganizationPopular

Zero friction. Full context. Competitive intelligence, M&A due diligence, and OSINT directly from your terminal. Uses Node.js/npx.

4.6K
49.5K
Instructions
View
kepano logo

Defuddle

kepano
CommunityPopular

Extract clean Markdown from HTML pages with Defuddle CLI.

3.5K
48.5K
Instructions
View

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇