Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/Rheinmir/setupnpx agentmods add skills/rheinmir/setup/dym-agent-desktop-sessionWrote 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/skills/rheinmir/setup/dym-agent-desktop-session)<a href="https://agentmods.dev/skills/rheinmir/setup/dym-agent-desktop-session"><img src="https://agentmods.dev/badge/skills/rheinmir/setup/dym-agent-desktop-session/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/skills/rheinmir/setup/dym-agent-desktop-session"><img src="https://agentmods.dev/badge/skills/rheinmir/setup/dym-agent-desktop-session.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 | $0.00071 | $0.01055 |
| Opus 5.5 | $0.00028 | $0.00422 |
| Sonnet 5.5 | $0.00014 | $0.00211 |
| Haiku 4.5 | $0.00007 | $0.00105 |
Grade A, and why
dym-agent-desktop-session 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 yesterday.
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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
dym-agent-desktop-session
When to use
- Một lượt thao tác desktop cần để lại biên lai:
/br qc,/fdk-poc, hay bất kỳ verdict nào phải cãi lại được bằng log chứ không bằng lời model. - Nhiều agent chạy song song trên cùng một máy macOS (đúng bối cảnh Orca worktree) — chuột là tài nguyên dùng chung, cần lease và cần biết cursor nào của agent nào.
- Chỉ chạy một lệnh đọc lẻ → bỏ qua skill này, session là thừa.
Steps
- Mở session một lần cho cả lượt việc (trace bật sẵn):
$AD session start --name br-frame-07 # → data.session_id, data.next
export AGENT_DESKTOP_SESSION=run-1788964205617-6007-0
Cần ảnh pre/post mỗi hành động để replay: session start --screenshots
(artifacts: full) — nhạy cảm, coi như screenshot, đừng đẩy vào repo.
-
Chạy việc bình thường — mọi lệnh trong scope tự ghi JSONL, không cần
--tracetừng lệnh. Segment nằm ở$AGENT_DESKTOP_HOME/sessions/<id>/trace/<pid>-<procTs>.jsonl. -
Đọc lại bằng chứng:
$AD trace show --limit 200 # JSON gọn cho agent đọc
$AD trace export --out ./scratchpad/run.html # HTML một file cho người xem
Đo thật: mỗi lệnh sinh command.start → sự kiện nghiệp vụ (vd snapshot.saved kèm
ref_count, snapshot_id) → command.end kèm duration_ms, ok. Có ts_ms và seq
đơn điệu theo tiến trình → ghép được thứ tự thật giữa nhiều tiến trình.
- Đóng và dọn:
$AD session end $AGENT_DESKTOP_SESSION
$AD session gc --ended --older-than 86400
- Nhiều agent song song (chưa kiểm chứng bằng chạy thật trong lượt này — theo doc của tool):
$AD session start --cursor --multi-agent # một session dùng chung
AGENT_DESKTOP_AGENT_ID=frame-07 $AD click "@s..:e5" # mỗi agent một ID ổn định
$AD cursor-overlay disable # tắt mọi cursor của session
Harness truyền cùng một session ID cho mọi subagent; mỗi subagent tự khai
AGENT_DESKTOP_AGENT_ID (hoặc --agent-id, cờ thắng env). Ba ID khác nhau → ba con trỏ
có màu/nhãn riêng. Thao tác UI bắt buộc có ID; quan sát/clipboard/quản trị thì không.
Lease tuần tự hoá truy cập vào con trỏ OS dùng chung.
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.
- yesterday First seen · 76 lines · 71 tokens per session scan A 7164db957f19
dym-agent-desktop-session is a skill published in the GitHub repository Rheinmir/setup (11 stars, last pushed 20d ago), licensed MIT. It adds 71 tokens to every session and 1,055 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-11-07.
dym-agent-desktop-session compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| dym-agent-desktop-sessionRheinmir/setup | Skill | 11 | 20d ago | MIT | A | 71 | Claude Code |
| interactive-loginyc-software/qm | Skill | 16k | today | MIT | A | 46 | Claude Code, Codex |
| pinchtab-mcppinchtab/pinchtab | Skill | 11k | 11d ago | MIT | A | 52 | Claude Code, Codex |
| azure-messaging-webpubsub-javamicrosoft/skills | Skill | 3.2k | yesterday | MIT | A | 43 | Claude Code |
| cua-driverdavidondrej/skills | Skill | 4.3k | yesterday | MIT | A | 68 | Claude Code, Codex |
| open-design-browsercompozy/compozy | Skill | 2.9k | today | MIT | A | 56 | Claude Code, Codex |
| southwestborski/travel-hacking-toolkit | Skill | 906 | 2d ago | MIT | A | 40 | Claude Code |