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.
git clone --depth 1 https://github.com/rakutentech/sorifyWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/rakutentech/sorify/recording)<a href="https://agentmods.dev/commands/rakutentech/sorify/recording"><img src="https://agentmods.dev/badge/commands/rakutentech/sorify/recording/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/rakutentech/sorify/recording"><img src="https://agentmods.dev/badge/commands/rakutentech/sorify/recording.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00104 | $0.02481 |
| Opus 5.5 | $0.00042 | $0.00992 |
| Sonnet 5.5 | $0.00021 | $0.00496 |
| Haiku 4.5 | $0.00010 | $0.00248 |
Grade C, and why
recording scanned grade C with 2 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 1mo 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.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
curl -fsSL https://raw.githubusercontent.com/rakutentech/sorify/master/mcp/install-mcp.sh | sh Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -fsSL https://raw.githubusercontent.com/rakutentech/sorify/master/mcp/install-mcp.sh | sh How it starts
The opening of the file, as written. The whole thing — 190 lines — stays where its author put it; the contents beside it link to each section on GitHub.
sorify:recording
Trigger
Invoked as /sorify:recording [session_id | "latest"].
Step 0: Verify the sorify-recorder MCP is installed
This command depends on the sorify-recorder MCP server (tools like
mcp__plugin_sorify_sorify-recorder__list_recordings). Unlike the sorify
MCP server (HTTP, wired up automatically when the plugin is installed), this
one is a local binary that does not install itself — check for it before
calling any recorder tool.
Check with /mcp → look for plugin:sorify:sorify-recorder. If it's missing,
or the first call to list_recordings fails to connect, abort with:
Error: sorify-recorder MCP is not installed.
Install it:
curl -fsSL https://raw.githubusercontent.com/rakutentech/sorify/master/mcp/install-mcp.sh | sh
This downloads the sorify-recorder-mcp binary for your OS/arch into
~/.sorify-bin/. Restart Claude Code (or run /mcp to reconnect) afterwards,
then also make sure the Sorify Recorder Chrome extension is loaded:
chrome://extensions/ → enable "Developer mode" → "Load unpacked" → select
the sorify repo's `extension/` folder.
Step 1: Resolve the recording
Call mcp__plugin_sorify_sorify-recorder__list_recordings — this returns
lightweight metadata only (session_id, label, path, event_count,
timestamps), never the recorded events themselves.
$ARGUMENTS is empty or "latest" → use the entry with the newest `mtime`
$ARGUMENTS is a session_id → match it exactly
If there are zero recordings → abort with:
Error: No recordings found. Record a session first — load the Sorify
Recorder extension unpacked, click the icon, Connect, Start recording.
If the resolved recording has event_count: 0 → abort with:
Error: Recording {session_id} has no captured events.
Keep the resolved path — this is the JSONL file on disk. Print which file
was resolved before going further:
Reading recording: {path} ({event_count} events, label: "{label}")
Step 2: Show interpretation and confirm before touching Sorify
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.
- 1mo ago Changed · +6 lines 01a8c1d3bc9a
- 1mo ago First seen · 184 lines · 104 tokens per session scan C 822f3d5ec1f4
recording is a command published in the GitHub repository rakutentech/sorify (50 stars, last pushed 2d ago), licensed Apache-2.0. It adds 104 tokens to every session and 2,481 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
mtestlocal
Run a real-browser E2E acceptance test against localhost (alias for /muggle-test-feature-local).
record
Record the browser while validating a flow, then judge it against a rubric.
expect
Diff-aware AI browser testing — reads the git diff, maps changes to affected pages via the route map, generates a targeted test plan, and executes it via agent-browser (Rust daemon + CDP, ARIA-tree-first) with pass/fail reporting. Use when testing UI changes, verifying PRs before merge, or running regression checks on…
record
Record a browser walkthrough of a URL using Antigravity (agy). Generates .webm video, screenshots, and a report. Auto-converts to MP4 if ffmpeg is available.
webapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
amazing-e2e-check
Run explicitly requested E2E, browser, device, or installed-deployment scenarios in an independent verification workflow.