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/kvker/ai-native-template/an-initnpx skills add kvker/ai-native-template --skill an-initgit clone --depth 1 https://github.com/kvker/ai-native-templateWrote 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/kvker/ai-native-template/an-init)<a href="https://agentmods.dev/skills/kvker/ai-native-template/an-init"><img src="https://agentmods.dev/badge/skills/kvker/ai-native-template/an-init.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 | $0.00059 | $0.01793 |
| Opus 5 | $0.00030 | $0.00897 |
| Sonnet 5 | $0.00012 | $0.00359 |
| Haiku 4.5 | $0.00006 | $0.00179 |
Grade C, and why
an-init scanned grade C 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 5d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
安装器先复制到 staging 并验证所有 Skill,再提交目标、原子移除待完成路由块,并将 `an-init` 退出发现路径。提交前失败会恢复根 AGENTS、已提交目标和 `an-init`;提交成功后的旧副本清理失败只报告 warning,不回滚已安装 Skills。不要手工执行 `mv ...; rm -rf ...`。 How it starts
The opening of the file, as written. The whole thing — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Native 上下文初始化
将 projects/ 中的材料纳入通用 AI Context 结构。初始化只整理事实和约定,不创建任务 Artifact,不预设行业。
一、前置检查
- 确认当前目录同时包含根
AGENTS.md、artifacts/index.json和.agents/skills/an-init/SKILL.md;否则停止,要求从工作区根目录重新执行。 - 检查
projects/下一级子目录,每个子目录视为独立工作单元。 projects/为空或不存在时,一次性询问是否从空模板开始;确认后创建目录并生成最小上下文。- 已有材料时直接继续,不要求用户填写能够从材料获得的信息。
- 独立工作单元默认可以用子代理并行扫描;环境不支持时由主代理顺序完成。
二、执行流程
工作区扫描 → 事实整理 → 上下文生成 → 动作探测 → 安装前检查 → 安装 Skills → 报告并重启
三、工作区扫描
从工作区根目录运行:
node .agents/skills/an-init/assets/skills/an-recipes/scripts/detect-recipes.mjs --root projects --write .agents/recipes.json
node .agents/skills/an-init/assets/skills/an-refresh/scripts/scan-projects.mjs --root projects --artifacts artifacts --format markdown
扫描输出中的 diagnostics 非空或 truncated: true 时,先针对目标材料补充检查并在报告中说明限制。扫描失败不得解释为“没有内容”。
扫描后按需阅读 README、索引、配置、注释和已有规范,提取:
- 目录、文件类型和主要材料。
- 已明确表达的用途、入口和外部约定。
- 声明的工具、脚本和可执行动作。
- 命名、文件组织和格式规律。
所有结论必须来自用户输入或工作区事实。可以基于证据归纳,但必须标注来源;置信度不足时写“待确认”,不得用领域常识补成已确认事实。
四、上下文生成
已有文件时保留仍有效的人工内容,只修改与事实冲突或已经过时的部分。
根 AGENTS.md
改写为运行期路由,至少包含:
- 已确认的工作区定位和工作单元路由。
- 目录职责和敏感文件约束。
- 必读规范。
- 指向
artifacts/index.json的任务状态路由。 $an-task、$an-task-split、$an-recipes、$an-refresh、$an-review、$an-archive的运行期 Skill 路由。$an-task触发规则:用户明确下达“开始吧”“开始做”“实现”“动手”“开工”等进入执行态的指令时,无论此前讨论多少轮,都必须先经$an-task分流;该规则写入运行期根 AGENTS。$an-task规模规则:小改动只允许不创建 Artifact,不允许跳过分流;预计变更涉及 3 个以上文件或跨前后端时至少使用 L2。
移除原模板阶段说明,但在安装完成前保留以下唯一待完成路由块;所有 $an-init 和 .agents/skills/an-init 引用都必须放在块内:
<!-- an-init-pending:start -->
初始化尚未完成时,使用 `$an-init` 继续;入口位于 `.agents/skills/an-init/`。
<!-- an-init-pending:end -->
根文件不复制活跃任务表或流程正文。安装器在所有运行期 Skill 提交成功后原子移除该块;失败或取消时该入口仍可重试。
根 AI 行为约束必须包含:
当任务需要访问某个目录时,先检查该目录及其相关父级路径中的 AGENTS.md(如存在),再读取或修改其中的文件。不要为此主动遍历无关目录。
README.md
将模板 README 改写为当前工作区的人类概览,至少说明用途、工作单元、目录、运行期 Skills 和启动方式。必须移除 <!-- ai-native-template-readme --> 标记;不要删除用户 README。
background
生成或更新:
background/overview.md:定位、已知目标、主要材料、待确认和来源。background/workspaces.md:每个工作单元的路径、类型、内容、动作和来源。background/AGENTS.md:稳定但可刷新、任务默认只读、已确认与待确认分离。
What ships with it
23 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.
- agents/openai.yaml 330 B
- assets/skills/an-archive/agents/openai.yaml 285 B
- assets/skills/an-archive/scripts/archive-task.mjs 13 KB runs code
- assets/skills/an-archive/SKILL.md.txt 3.6 KB
- assets/skills/an-recipes/agents/openai.yaml 275 B
- assets/skills/an-recipes/scripts/detect-recipes.mjs 18 KB runs code
- assets/skills/an-recipes/SKILL.md.txt 2.5 KB
- assets/skills/an-refresh/agents/openai.yaml 279 B
- assets/skills/an-refresh/scripts/scan-projects.mjs 23 KB runs code
- assets/skills/an-refresh/SKILL.md.txt 3.0 KB
- assets/skills/an-review/agents/openai.yaml 268 B
- assets/skills/an-review/scripts/review-task.mjs 21 KB runs code
- assets/skills/an-review/SKILL.md.txt 2.9 KB
- assets/skills/an-task-split/agents/openai.yaml 286 B
- assets/skills/an-task-split/SKILL.md.txt 2.8 KB
- assets/skills/an-task/agents/openai.yaml 335 B
- assets/skills/an-task/SKILL.md.txt 7.5 KB
- scripts/install-runtime.mjs 11 KB runs code
- tests/archive-task.test.mjs 11 KB runs code
- tests/install-runtime.test.mjs 6.2 KB runs code
- tests/review-task.test.mjs 11 KB runs code
- tests/scanners.test.mjs 20 KB runs code
- tests/template-contract.test.mjs 5.0 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.
- 5d ago First seen · 153 lines · 59 tokens per session scan C e6d147716a13
an-init is a skill published in the GitHub repository kvker/ai-native-template (101 stars, last pushed 24d ago), licensed MIT. It adds 59 tokens to every session and 1,793 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
media-ingest
Ingest video, audio, PDF, book, screenshot, and GitHub repo content into the brain. Multi-format handling with entity extraction and backlink propagation. Covers video-ingest, youtube-ingest, and book-ingest subtypes.
mem0-oss-to-platform
Plan and then execute a migration of a project from the mem0 open-source / self-hosted SDK (the local Memory class) to the mem0 Platform / hosted / managed SDK (the MemoryClient class). Use this whenever a developer wants to move, switch, or migrate their mem0 usage off OSS/self-hosted to the hosted API — e.g.…
Cortex
Operate Cortex, the LifeOS memory system — the typed Knowledge Archive (People, Companies, Ideas, Research with typed related: links) plus recall of prior work sessions, ISAs, and conversations. Search, add, harvest, develop, ingest, distill, graph-navigate, recall. USE WHEN cortex, knowledge, knowledge base, search…
agent-memory
../../../engineering/agent-memory/skills/agent-memory/SKILL.md.
memory
Use when the user asks to remember, recall, forget, update, search, or inspect durable OpenSquilla memory, including profile facts in USER.md and long-term notes in MEMORY.md or memory//.md.
ha-data-stores
Map of Hope Agent's local data stores and safe read-only query workflow. Use when the user asks where Hope Agent stores data, wants to inspect sessions/messages/memory/logs/background jobs/knowledge indexes/settings, asks the model to query local app data, or debugging requires checking persisted state. Trigger…