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.
npx agentmods add commands/j-ronk/backchannel/joingit clone --depth 1 https://github.com/j-ronk/backchannelWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00009 | $0.00255 |
| Opus 5 | $0.00005 | $0.00128 |
| Sonnet 5 | $0.00002 | $0.00051 |
| Haiku 4.5 | $0.00001 | $0.00026 |
Grade A, and why
join 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.
What it actually says
Join a shared backchannel room. The user's input — a room link (starts with https:// and contains a #… fragment that carries the room key), followed by a REQUIRED display name — is:
$ARGUMENTS
A display name is required. If the user gave only a link and no name, ask them for a display name before joining (don't invent one).
Do NOT use inline ! execution for this (the link's # fragment gets mangled by shell/argument substitution). Instead, use the Bash tool to run the backchannel CLI, passing the entire link as a single quoted argument (everything from https:// through the end of the #k=… fragment, with no truncation) and the display name via --name:
"${CLAUDE_PLUGIN_ROOT}/bin/backchannel" join "<full link including #k=… fragment>" --name "<display name>"
Then show the user the CLI's confirmation message verbatim.
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.
- 2d ago First seen · 19 lines · 9 tokens per session scan A e63a68df95f6
join is a command published in the GitHub repository j-ronk/backchannel (1 stars, last pushed 1mo ago), licensed MIT. It adds 9 tokens to every session and 255 once invoked, about $0.0000 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 2026-08-31.
Other commands, from other repositories
AGENTS
CLAUDE.md is a symlink to this file. Edit AGENTS.md only.
recap
Get a summary of everything you learned this session — concepts, quizzes, and progress.
status
Show current session collaboration status including active sessions, claims, and your session info.
aws-spec
List, inspect, or validate per-feature spec folders under .claude/specs/ / produced by the claude-aws-architect orchestrator. Runs deterministic and runtime gates against requirements.md, design.md, tasks.md, contracts/, and diagrams.d2.
board
Create an Overboard Studio board from your current conversation. Usage: /overboard [type] [description]. Types: plan, diagram, brainstorm, meeting, strategy, flowchart, kanban, timeline, mindmap, retrospective, swot. Example: /overboard plan Launch mobile app by Q3.
setup
Set up the Overboard Studio MCP connection for creating boards from Claude Code.