role

A command that assigns a named role to the current coding-agent session, such as frontend, backend, or tester. Other sessions can then send messages to that role through the agent bus.

In plain words
What is it for?
It is for binding this session to a supplied role or suggesting possible roles from the conversation when no role is given.
Why use it?
It helps teams of agent sessions direct questions and work to the right session instead of addressing them ambiguously.

Command

Part of the claude-plugin plugin — 1 skill, 7 commands, 1 agent, 1 MCP server shipped together

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 commands/mercurialsolo/claudectl/role
Clone the repo
git clone --depth 1 https://github.com/mercurialsolo/claudectl

Or install claude-plugin, the plugin that ships this one along with the rest of its 1 skill, 7 commands, 1 agent, 1 MCP server.

Per session 33 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 498 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 $0.00033 $0.00498
Opus 5 $0.00016 $0.00249
Sonnet 5 $0.00007 $0.00100
Haiku 4.5 $0.00003 $0.00050

Measured 3d ago against content hash fd6c4e3bfc6a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

role 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 3d 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.

claude-plugin/commands/role.md · 34 lines

What it actually says

Set this session's agent-bus role so peers can send it directed messages with claudectl bus send <role> …. The binding is PID-keyed — it follows this exact Claude Code process, so re-runs in the same project don't get confused by other sessions sharing the cwd.

Common role names: spec, frontend, backend, data-analyst, tester, infra, reviewer, planner. Pick whatever maps to what this session will actually own.

What to do

  1. The user supplied a role name as {{args}}. If it's empty, run claudectl bus role suggest --json --top 3 to fetch transcript-derived candidates and propose them; if even that comes back empty, ask the user what to name the role and stop until they reply.

  2. Run:

    claudectl bus role bind --self {{args}}
    

    The --self flag tells the CLI to walk the ancestor process chain, find this Claude Code process, and bind the role to that pid + the current working directory. No need to look up the pid yourself.

  3. On success the command prints bound role <name> -> <cwd> (pid=<pid>). Echo that confirmation back to the user.

  4. If the command errors with "could not find a Claude Code process in the ancestor chain", the plugin was invoked outside the normal claude execution context. Tell the user to try again from inside a Claude session, or pass an explicit pid via claudectl bus role bind <name> <cwd> --pid <pid>.

  5. If the user wants to verify the binding, suggest claudectl bus role list or claudectl bus whoami --json.

Why pid-binding

Cwd-only bindings get ambiguous when two Claude sessions run in the same folder (e.g. worktrees, or one session in /repo and another in /repo/sub). The pid binding picks the specific session you ran /role from, so directed messages land where you expect.

See issues #307 (pid binding) and #310 (this slash command).

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. 3d ago First seen · 34 lines · 33 tokens per session scan A fd6c4e3bfc6a

Subscribe to this mod's changes

role is a command published in the GitHub repository mercurialsolo/claudectl (197 stars, last pushed 1mo ago), licensed MIT. It adds 33 tokens to every session and 498 once invoked, about $0.0002 per session on Opus 5. 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.