Echo Agent is a self-hosted, long-running AI agent that connects language models, tools, memory, permissions, and messaging channels in one system. Individuals and teams use it for private automation that retains context across sessions, develops skills, schedules tasks, and requires approval for high-risk actions.
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 skills add fuyuxiang/echo-agent --skill remindergit clone --depth 1 https://github.com/fuyuxiang/echo-agentWrote 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/fuyuxiang/echo-agent/reminder)<a href="https://agentmods.dev/skills/fuyuxiang/echo-agent/reminder"><img src="https://agentmods.dev/badge/skills/fuyuxiang/echo-agent/reminder/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/fuyuxiang/echo-agent/reminder"><img src="https://agentmods.dev/badge/skills/fuyuxiang/echo-agent/reminder.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00023 | $0.01126 |
| Opus 5.5 | $0.00009 | $0.00450 |
| Sonnet 5.5 | $0.00005 | $0.00225 |
| Haiku 4.5 | $0.00002 | $0.00113 |
Grade A, and why
reminder 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 — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Reminder
Set timed reminders and manage todos.
IMPORTANT: how a reminder actually fires
A row in reminders.db is just a note — nothing scans it or delivers it. To make
a reminder that actually fires and gets delivered, you MUST create a scheduled
job with the cronjob tool, which persists to the scheduler and is delivered
through the active channel by Echo Agent's scheduler service:
cronjob(action="create", name="weekly-report",
schedule="0 9 * * 1", command="提醒你写周报")
The reminder_store.py script is local bookkeeping with no delivery path. Use cronjob for recurring delivery and optionally record a local note. Do not delegate cron authorization to a background worker: create and approve the job in the interactive parent first.
The public cronjob tool currently creates recurring CRON jobs only. Its API does not expose the scheduler's internal one-shot at_ms trigger. Do not promise a one-off reminder by turning a date into a repeating day-of-month expression; explain that boundary before creating anything.
Jobs that produce a file (audio/image/document)
When the scheduled command generates an artifact the user must receive (e.g. a
voice briefing), make the delivery happen inside the same tool call — do not
assume a later step will send it. For audio use text_to_speech(..., deliver=true)
(see the tts-voice skill); for other files call send_file explicitly with the
target channel/chat. An unattended cron run may end right after producing the
file, so "generate then hope it gets sent" silently drops the artifact.
After creating a job: confirm, then stop
Once cronjob(action="create") returns, you are done. Send the user ONE
confirmation with the concrete facts and end the turn:
- job id, the schedule, and the next fire time (from the create result)
- if a delivery target was inferred, say so; if the create result carried the "⚠️ 无法确定投递目标" warning, relay it and ask for target_channel/target_chat_id.
Do NOT try to "prove delivery works" by exercising the agent's own plumbing —
reading gateway/server.py, a2a/server.py, or curling internal endpoints like
/v1/chat or /api/v1/message is not verification, it's a rabbit hole that
burns the turn and leaves the user with no reply. If you genuinely need to sanity
-check the schedule, use cronjob(action="list"); trigger a real run at most
once and only if the user asked. The scheduled job itself is the delivery test —
it will fire on schedule.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 Changed · -2 lines 692153d7eeb7
- 1mo ago First seen · 116 lines · 23 tokens per session scan A 6c30ec95c566
reminder is a skill published in the GitHub repository fuyuxiang/echo-agent (1,035 stars, last pushed today), licensed MIT. It adds 23 tokens to every session and 1,126 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
agent-state-bar
A design guide for an agent status bar: a short, automatically updated summary of progress, remaining tasks, tool counts, errors, and system state placed near the latest model input.
knowledge-org
A guide to organising large bodies of knowledge so an AI system can navigate relationships, summaries, and layers instead of searching only flat text chunks. It covers tree indexes, knowledge graphs, file-like structures, incremental updates, and long-term memory layouts.
memory-system
A guide to giving an AI agent persistent, user-specific memory across conversations. It covers what to remember, how to store it, how to update or compress it, and when to use memory instead of a shared knowledge base.
context-compression
A guide to shortening the information an agent carries between steps. It explains how to replace large tool outputs and old conversation details with compact, useful summaries while preserving important decisions and constraints.
dingtalk-message
A tool for sending DingTalk messages through either a group Webhook robot or an internal enterprise robot. DingTalk is a workplace messaging platform.
computer-history
Answer questions about what the user recently did on their computer — who contacted them, what they were working on, where they left off — by reading the local Computer History summaries and raw event streams.