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/zilliztech/mfs/open-tagnpx skills add zilliztech/mfs --skill open-taggit clone --depth 1 https://github.com/zilliztech/mfsWrote 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/zilliztech/mfs/open-tag)<a href="https://agentmods.dev/skills/zilliztech/mfs/open-tag"><img src="https://agentmods.dev/badge/skills/zilliztech/mfs/open-tag.svg" alt="Measured on agentmods" 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 | $0.00096 | $0.01622 |
| Opus 5 | $0.00048 | $0.00811 |
| Sonnet 5 | $0.00019 | $0.00324 |
| Haiku 4.5 | $0.00010 | $0.00162 |
Grade A, and why
open-tag-admin scanned grade A with 1 finding 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 6d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
`curl -s 127.0.0.1:13619/healthz`. 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.
Open Tag (admin)
This skill is the control console for an Open Tag deployment. Use it for the first-time setup and for ongoing operation alike: inspect the live bot, change the backend or permitted scopes, add a new data source, move the bot to another Slack channel, or debug a run.
Keep the architecture generic:
- Brain: the selected CLI agent backend —
claude -p(Claude Code) orcodex exec(Codex). - Memory: MFS-indexed, operator-authorized context such as Slack history, repositories, docs, issues, databases, or object stores.
- Tools: MFS connectors for external read/search plus any explicit tools the backend is allowed to use in the workspace.
The user-facing flow is:
- Configure MFS sources and allowed scopes.
- Configure a Slack app with Socket Mode and invite it to a sandbox channel.
- Start the Open Tag bridge.
- Move to Slack and tag the bot in a thread.
- Let the bridge invoke the selected backend with thread context and scoped MFS helper scripts for permitted external context.
This skill does not call a model API directly. Model access, tool access, and write permissions come from the selected CLI agent backend.
Prerequisites
Open Tag is a thin layer on top of a running MFS server with at least one indexed source. Confirm these before any Slack work:
- MFS server installed and running.
uv tool install mfs-server, thenmfs-server run(binds127.0.0.1:13619). Check withcurl -s 127.0.0.1:13619/healthz. - At least one data source indexed. Open Tag only consumes already-indexed
scopes as Memory — it does not configure connectors itself. Use the
mfs-ingest skill (Codex:
$mfs-ingest) to add a source; see "Adding data sources" below.
opentag_doctor.py fails fast with a hint if either is missing.
Bot name convention
The Slack display name is whatever you call the Slack app — Open Tag's code
strips the mention regardless. Recommended convention, so it reads like the
official @Claude tag:
What ships with it
10 files 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.
- references/backends.md 2.1 KB
- references/memory.md 1.2 KB
- references/runtime-agent.md 3.4 KB
- references/slack-adapter.md 8.4 KB
- scripts/mfs_cat.py 2.2 KB runs code
- scripts/mfs_search.py 3.6 KB runs code
- scripts/opentag_agent.py 6.9 KB runs code
- scripts/opentag_doctor.py 6.3 KB runs code
- scripts/opentag_memory.py 3.2 KB runs code
- scripts/slack_socket_agent.py 5.6 KB runs code
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.
- 6d ago First seen · 143 lines · 96 tokens per session scan A 894e7aac5184
open-tag-admin is a skill published in the GitHub repository zilliztech/mfs (135 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 96 tokens to every session and 1,622 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
memx
Use when an agent needs to recall, save, inspect, or audit memX memories through the bundled MCP tools.
ai-style
当任务是用中文撰写或改写面向读者的文案（产品发布稿、公众号文章、邮件、README 等）， 或用户反馈文字「AI 味太重」「不像人写的」时，加载本 Skill。.
triage
你是当前任务的分诊协调者。先识别用户的全部目标、顺序依赖和验收条件，再按 “事实检索 → 计算/执行 → 写作”顺序逐步请求切换到需要的专业能力。不要替专业 能力完成它的工作，也不要在信息缺失时臆造结果。.
writing
将共享历史中的已验证事实和计算结果整理成符合受众、格式与长度约束的成稿。.
together-ai-inference
Serverless inference, fine-tuning, embeddings, image generation, and batch processing on 200+ open-source models via an OpenAI-compatible API. Use when you need fast, cost-effective access to open-source LLMs without managing infrastructure.
coding
编写并运行 Python 代码，验证脚本逻辑和输出。.