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/gnosil/semantix/agent-skillnpx skills add Gnosil/semantix --skill agent-skillgit clone --depth 1 https://github.com/Gnosil/semantixWrote 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/gnosil/semantix/agent-skill)<a href="https://agentmods.dev/skills/gnosil/semantix/agent-skill"><img src="https://agentmods.dev/badge/skills/gnosil/semantix/agent-skill.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 | $0.00047 | $0.02370 |
| Opus 5 | $0.00023 | $0.01185 |
| Sonnet 5 | $0.00009 | $0.00474 |
| Haiku 4.5 | $0.00005 | $0.00237 |
Grade A, and why
semantix 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 4d 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.
command -v curl || echo "need curl" How it starts
The opening of the file, as written. The whole thing — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Semantix — 记忆中间件 Skill
给你的智能体装上用户级记忆：记住用户偏好、业务流程、总结办案/工作经验， 跨会话自动复用。你的 agent 不需要改造内核——装一个二进制 + 两个工具即可。
执行模式（agent 自助接入，逐条执行）
本 skill 设计为由客户的 agent 直接执行完成接入。按顺序执行以下步骤； 每步有通过标准；失败进入「故障处理」。全部通过 = 接入完成。
STEP 0 — 环境检查
uname -s && uname -m # 通过标准：darwin 或 linux；amd64 或 arm64
command -v curl || echo "need curl"
STEP 1 — 安装
bash <(curl -sL https://raw.githubusercontent.com/Gnosil/semantix/main/agent-skill/scripts/install.sh)
通过标准：输出 == installed: 且无报错。
（离线/内网客户：手动下载 release 包，跳过下载步骤后继续。）
STEP 2 — 配置
semantix lookup --help >/dev/null && echo ok # 通过标准：输出 ok
mkdir -p ~/.semantix/sessions
默认零配置可用；按需复制 agent-skill/config/semantix.example.toml 调整。
STEP 3 — 选择集成方式（决策树）
| 你的 agent 底座 | 走哪条路径 |
|---|---|
| semantix-agent（本仓 harness/，随发布包分发） | 内置集成已就绪：semantix-agent.toml 中 [semantix] enabled=true + semantix_lookup 工具（零步骤） |
| LangChain | docs/reports/langchain-middleware.md 双挂点示例（消息改写 + 会话提取） |
| Claude Code / 工具注册类 | 本 skill tools/semantix-lookup.md 的 schema 注册 semantix_lookup（+可选 semantix_inject） |
| 任意自定义 agent | 会话旁路三方式（hooks/session-bypass.md）：导出 / 事件旁路 / 直接调用 |
STEP 3.5 — Claude Code 接第三方 GLM 端点时的前缀卫生（可选，强烈推荐）
Claude Code ≥2.1.36 会在 system prompt 最前端注入逐请求变化的计费头
（x-anthropic-billing-header 的 cch 字段）。官方服务端会剥离它；第三方 GLM 端点
（腾讯云 TokenHub 等）不剥离——前缀从第一个 token 起失配，隐式缓存全灭
（社区实测命中差 133 倍；对照 TokenHub 官方 Claude Code 接入页推荐的环境变量组）。
两道防线任开其一，同开更稳：
export CLAUDE_CODE_ATTRIBUTION_HEADER=0 # 客户端不再注入计费头（直连第三方端点时唯一有效的一道）
- 经 semantix-gateway 转发时：网关净化中间件默认剥离该计费头（
[sanitize] strip_attribution，见deploy/semantix-gateway.toml.example），未设环境变量也受保护； - 直连第三方端点（不过网关）时：只有上面的环境变量能救。
STEP 4 — 闭环自测（必须通过）
bash <(curl -sL https://raw.githubusercontent.com/Gnosil/semantix/main/agent-skill/scripts/selftest.sh)
通过标准：输出 SELFTEST PASS。自测内容：
提取 demo 会话 → lookup 命中（zone=hit）→ inject 返回注入块 → 记忆闭环成立。
STEP 5 — 接入完成
- 会话结束：
extract --input <会话> --scope user --project <业务域>（沉淀） - 新任务：
inject --query <任务> --scope user（注入）或lookup（检索） - 定期：
verify --session <目录> --scope user（质量评估）
What ships with it
5 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.
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.
- 4d ago First seen · 177 lines · 47 tokens per session scan A ca97cc193118
semantix is a skill published in the GitHub repository Gnosil/semantix (330 stars, last pushed today), licensed MIT. It adds 47 tokens to every session and 2,370 once invoked, about $0.0002 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
faq-mine
Mine docs/faq.md from README.md, docs/.md, and the pi-hermes memory stores. Dispatches @fast subagents per source, dedupes against the existing FAQ, and merges entries in caveman style. Use when asked to "build / regenerate / extend the FAQ", "mine docs into FAQ", "mine hermes memory into FAQ", "surface runtime…
distill-session-knowledge
Offline-mine this project's pi session JSONL logs into reusable, verified knowledge: extracts faults, decisions, corrections, procedures and docs, promotes only recurring patterns, and routes artifacts into skillmanage, memory and docs. Use on "mine my sessions", "distill session knowledge", "extract lessons from…
context-engineering
Use when designing, reviewing, or debugging how an agent's context window gets filled, pruned, or shared — choosing what loads at boot versus on demand, sizing an install or an always-loaded file, fixing an agent that drifts, repeats itself, or forgets constraints mid-task, planning compaction or summarization…
design-memory
Use at project start and completion to OBSERVE and record how the user designs — the decisions they make, the styles they reach for, their habits and inclinations across projects. This is a descriptive record (a mirror), NOT a controller — it is never fed back to steer the work. Per-project direction comes from…
context-optimizer
Optimize token usage and context management. Use when sessions feel slow, context is degraded, or you're running out of budget.
deep-clean
Full-spectrum consolidation of AI agent configuration files. Goes beyond memory-only dream skills: audits and optimizes context files (AGENTS.md/AGENTS.md/GEMINI.md/.cursorrules), rules, skills, and memory. Detects stale references, dead file paths, duplicated rules, stack mismatches, contradictions, vague directives…