commands

A system for deciding how slash commands in an agent conversation should be handled. It determines whether a command controls the conversation, runs an agent action, or is passed unchanged to the underlying agent.

In plain words
What is it for?
Use it to define and implement slash-command dispatch inside a conversation session. It helps route commands such as controls or native agent commands to the correct handler.
Why use it?
It prevents command handling from becoming mixed up between the conversation controller and the agent that performs the work. Reserved control commands take priority over agent commands.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/longbkit/clisbot/commands
Clone the repo
git clone --depth 1 https://github.com/longbkit/clisbot
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,375 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.01375
Opus 5 $0.00000 $0.00687
Sonnet 5 $0.00000 $0.00275
Haiku 4.5 $0.00000 $0.00137

Measured 2d ago against content hash f31b6115402a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

commands scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

docs/features/agents/commands.md · 148 lines

How it starts

The opening of the file, as written. The whole thing — 148 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Commands

Purpose

This document defines agent-scoped command dispatch inside a conversation.

It covers slash-prefixed commands and configurable shorthand command forms that are interpreted before a prompt is sent to the runner.

Boundary Rule

Agent command dispatch belongs to agents.

Why:

  • it decides whether inbound text is treated as agent control or normal agent input
  • it is scoped to one conversation session key inside an agent
  • it must remain meaningful even if the concrete runner changes

It does not belong to runners, because runners should receive already-decided input.

It does not belong to control, because these commands are used inside a user conversation rather than an operator-only surface.

Dispatch Rule

When a message starts with /:

  1. check whether it matches a reserved control slash command
  2. if it matches, execute the control command immediately
  3. if it matches an agent-reserved execution command, execute that command
  4. otherwise, forward the slash command unchanged to the agent as native runner input

Control slash commands always have higher priority than native agent slash commands.

When a message starts with a configured bash shortcut such as !:

  1. treat the remaining text as an agent-scoped bash command
  2. execute it in the current agent workspace

Current Control Slash Commands

  • /start: show onboarding help for the current surface
  • /help: show the available control slash commands
  • /status: show the current route status and operator setup hints
  • /whoami: return the current platform sender and route identity plus the current session id view for the active conversation
  • /transcript: return the current full conversation session transcript
  • /stop: send Escape to interrupt current processing in the current conversation session, clear clisbot's active-run state, and let queued prompts continue
  • /new: start a new session for the current routed conversation and save the new sessionId
  • /nudge: send one extra Enter to the current tmux session without resending the prompt body
  • /queue <message> or \q <message>: create one durable queued prompt for the current session; attachments sent with the command stay attached to the queued prompt
  • /queue list: show pending queued prompts for the current session
  • /queue clear: clear pending queued prompts for the current session without interrupting a running prompt
  • /followup status
  • /followup auto
  • /followup mention-only or /mention
  • /followup mention-only channel or /mention channel
  • /followup mention-only all or /mention all
  • /followup pause or /pause
  • /followup resume or /resume

Read the full file on GitHub · 148 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 148 lines · 0 tokens per session scan A f31b6115402a

Subscribe to this mod's changes

commands is an agent published in the GitHub repository longbkit/clisbot (114 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,375 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.