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 skills/slaveoftime/open-relay/olynpx skills add slaveOftime/open-relay --skill olygit clone --depth 1 https://github.com/slaveOftime/open-relayWhat 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 | $0.00037 | $0.00938 |
| Opus 5 | $0.00018 | $0.00469 |
| Sonnet 5 | $0.00007 | $0.00188 |
| Haiku 4.5 | $0.00004 | $0.00094 |
Grade A, and why
oly 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.
How it starts
The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
When to use
Use oly start instead of a direct terminal invocation when ANY of these apply:
- The command may prompt for input later.
- The session must survive terminal closes, disconnects, or agent handoff.
- Logs, replay, or auditability matter.
- Another human or agent may need to resume or inspect the work.
Do not use oly for short, non-interactive commands — a normal terminal is simpler.
Principles
- Supervisor mindset. Start the job, monitor it, intervene only when needed.
- Small tails first. Use
--tail 40; expand only when context is insufficient. - Fewer polls, longer waits. Set
--timeoutto match the expected next checkpoint — reduces churn and token cost. - Machine-readable listing. Use
oly ls --json --status runningfor scripting or structured decisions.
Workflow
1) Start
oly start --title "task 1" --cwd /path/to/dir --detach the_cmd --arg1 --arg2
| Flag | Purpose |
|---|---|
--detach |
Return immediately with the session ID. |
--disable-notifications |
Suppress notifications when you will supervise yourself. |
--node <name> |
Run on a connected secondary node. |
2) Monitor
oly logs <ID> --tail 40 --no-truncate --wait-for-prompt --timeout 10s
--wait-for-prompt— blocks until the session likely needs input or timeout expires.--timeout— accepts250ms,10s,5m,1h(default5m). Shorten for fast tasks; lengthen for slow ones.- Start with
--tail 40; increase only when recent context is insufficient.
3) Send input
oly send <ID> "hello world!" key:enter
oly send <ID> oly-clipboard
oly send <ID> "cat " oly-file:/path/to/file key:enter
- Arguments are sent left-to-right. Plain text is literal; special keys use
key:prefix. - Keys:
key:enter,key:ctrl+c,key:alt+x,key:shift+tab,key:up,key:down. - Raw hex:
key:hex:.... Piped stdin is supported when no positional chunks are given. - Clipboard content uses
oly-clipboard. Local file upload usesoly-file:<path>. This is normally used for remote node sessions.
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.
- 3d ago First seen · 100 lines · 37 tokens per session scan A 45a91119ff61
oly is a skill published in the GitHub repository slaveOftime/open-relay (93 stars, last pushed 7d ago), licensed MIT. It adds 37 tokens to every session and 938 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.
Other skills, from other repositories
start
Session initialization and lifecycle management: bootstraps session context, organizes files, generates CLAUDE.md, manages soul purpose lifecycle with completion protocol and active context harvesting. Use when user says /start, /init, bootstrap session, initialize session, or organize project.
s-compact
Session handoff for the next session — write-side pair of /s-continue. Triggers on "s-compact", "session handoff", "handoff prompt", "hand off", "wrap up the session", "prep for next session". Distills what /s-continue cannot recover: subagent findings, tool-output numbers, and decisions that never entered the…
stop
Gracefully close, pause, or verify a session: harvest promotable content, save progress, archive soul purpose, settle AtlasCoin bounty. Use when user says /stop, wrap up, done for the day, finishing up, close session, pause, verify, or end session.
codex-sessions-manager
Use this skill when the user wants to inspect, search, export, verify, clean up, delete, restore, or purge local Codex sessions stored under /.codex.
cadence-handoff
整理当前 session 讨论内容到 cadence 档案(v0.4:触发整 阶段（ε 整合）兜底 + 写 sha1 书签式 snapshot,15-30 行)。用户说 'handoff'/'整理 session'/'换新 session' 或跑 /cadence-handoff 时触发。.
obsidian-ingest
Превращает markdown-источник (статья, вебклип, транскрипт, конспект) в знания базы: литературная заметка, атомарки, связи. Срабатывай на «добавить в базу знаний», «переработать статью», «обработать конспект», «ингестировать», «положить в SecondBrain».