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/dotdevdotdev/agentwire-dev/handoffgit clone --depth 1 https://github.com/dotdevdotdev/agentwire-devWhat 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 | $0.00022 | $0.01548 |
| Opus 5 | $0.00011 | $0.00774 |
| Sonnet 5 | $0.00004 | $0.00310 |
| Haiku 4.5 | $0.00002 | $0.00155 |
Grade A, and why
handoff 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.
How it starts
The opening of the file, as written. The whole thing — 143 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/handoff
You are about to compile this conversation into a shareable handoff bundle. The output is two artifacts:
ai-handoff.md— XML-tagged markdown a teammate can paste into another LLM (Opus 4.7) to roughly continue the conversation.show-the-story.html— single-file presentation a human can open in a browser.
You are uniquely positioned to do this well: you already have full conversation context. Don't ask the user clarifying questions unless something is genuinely ambiguous — distill what you know.
Workflow
1. Initialize the bundle
Call the MCP tool to create the bundle dir and pre-fill the template with git state and the CLAUDE.md/rules/memory chain:
mcp__agentwire__handoff_init(title="<short slug>")
The tool returns a bundle_dir and an ai_handoff_path. The pre-filled template at ai_handoff_path already has:
<metadata>populated with cwd, branch, commit, repo url<instructions>populated with the full CLAUDE.md chain (verbatim)<project_state>populated withgit status,git log, and uncommitted diff
You must not touch the <instructions> section unless redacting something sensitive — it's what makes the bundle portable across machines.
2. Read the pre-filled template
Use Read on ai_handoff_path to see exactly what got pre-filled.
3. Fill in the rest
Use Write (or Edit) on ai_handoff_path to replace every {{ ... }} placeholder with real content. Specifically:
<title>
One short line summarizing the session — what it was about, not what was decided. Roughly 4-8 words.
<metadata> extras
Fill started_at / ended_at if you can estimate them, user_identity (if mentioned), and mcp_servers (the ones actually used in this conversation).
<environment>
What the receiver can't see from cwd alone. Active panes, channels (Slack/Discord/email), scheduler state if relevant. If nothing notable, write a one-line "no special environment" note.
<conversation_summary>
This is the most load-bearing section. Be dense and structured.
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 · 143 lines · 22 tokens per session scan A 1a8c9a8a62bf
handoff is a command published in the GitHub repository dotdevdotdev/agentwire-dev (21 stars, last pushed 17d ago), licensed Apache-2.0. It adds 22 tokens to every session and 1,548 once invoked, about $0.0001 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.
Other commands, from other repositories
agentlas
Agentlas commands — one entry point for network, build, storm, call, and the rest.
agentlas-login
Sign this machine into Agentlas (opens the browser sign-in window).
barebrowse
Browser automation using the user's real browser with real cookies. Handles consent walls, login sessions, and bot detection automatically.
live-canvas
Conduct design interviews, generate UI variations, and collect live click-to-annotate feedback that streams into the session so edits land without leaving the browser. Use when the user wants rapid iterative UI refinement, not just batched feedback.
run-test
Execute AI agent test script in real browser.
playwright-auth
Use when the user needs to log into a website for browser automation, save auth cookies, or sync browser profiles to cloud.