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 instructions/ddpie/lark-mcp-on-agentcore/agents-mdgit clone --depth 1 https://github.com/ddpie/lark-mcp-on-agentcoreWrote 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/instructions/ddpie/lark-mcp-on-agentcore/agents-md)<a href="https://agentmods.dev/instructions/ddpie/lark-mcp-on-agentcore/agents-md"><img src="https://agentmods.dev/badge/instructions/ddpie/lark-mcp-on-agentcore/agents-md.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 | $0.01843 | $0.01843 |
| Opus 5 | $0.00922 | $0.00922 |
| Sonnet 5 | $0.00369 | $0.00369 |
| Haiku 4.5 | $0.00184 | $0.00184 |
Grade A, and why
lark-mcp-on-agentcore AGENTS.md 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 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.
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 — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Guidance for AI coding agents working in this repo. This file is the single
source of truth for agent conventions; it links out rather than duplicating.
Human onboarding docs live in docs/ (bilingual _en/_zh).
Project overview
Hosted remote-MCP service on AWS Bedrock AgentCore that wraps lark-cli so
remote-MCP clients can call Feishu's 2500+ APIs via 200+ tools. Identity is
user-only (each user calls as themselves; no bot identity, no event stream).
Languages: TypeScript (lambda/, infra/), JavaScript (docker/, no build
step), Bash (scripts/). Node 20, ARM64-only container. AWS CDK builds the image,
IAM, Lambdas, and alarms; the AgentCore Runtime itself (its env vars, idle
timeout, request-header allowlist) is provisioned by scripts/deploy.sh via boto3,
not CDK — see docs/agent/architecture.md.
Each MCP session runs in its own AgentCore microVM — see
docs/agent/architecture.md for the request lifecycle and concurrency model.
Setup & commands
npm install # root deps (lambda + docker tooling)
cd infra && npm install # CDK deps
./scripts/test.sh # default offline: unit + typecheck + lint
./scripts/test.sh --unit # vitest only
./scripts/test.sh --full # everything incl. smoke/mcp-protocol/audit/e2e (needs Docker/AWS)
npm run lint # eslint (no Prettier); npm run lint:fix to autofix
npm run knip # dead-code scan
./scripts/deploy.sh # interactive deploy (re-run uses saved config)
./scripts/ops.sh status # operations toolkit (status/list-users/list-apps/rebuild-registry/rename/revoke/refresh-all/logs/rotate-secret/destroy)
./scripts/teardown.sh # destroy all resources
./scripts/upgrade.sh --all # multi-app: canary the default app, then upgrade the rest
# Multi-Feishu-app (one AWS account/region hosts N apps; design in .claude/specs/2026-06-07-multi-app-*):
# Bare `deploy.sh` at a TTY shows an app picker (default / existing / new); --app skips it.
./scripts/deploy.sh --app <slug> --alias "<name>" # deploy/onboard a named app (slug = a-z0-9-, ≤20)
./scripts/ops.sh --app <slug> status # operate one app (default app omits --app)
./scripts/teardown.sh --app <slug> # tear down one app (shared WAF kept if others remain)
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 · 139 lines · 1,843 tokens per session scan A b3b09c348251
lark-mcp-on-agentcore AGENTS.md is an instructions file published in the GitHub repository ddpie/lark-mcp-on-agentcore (8 stars, last pushed 7d ago), licensed MIT. It adds 1,843 tokens to every session, about $0.0092 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 2026-08-31.
Other instructions, from other repositories
botmux CLAUDE.md
Claude Code instructions for deepcoldy/botmux, covering botmux, 构建 & 运行, bun 开发链路, worktree 的 nodemodules：共享还是独立（改前必读） and 编译态（单文件二进制）注意.
openlark AGENTS.md
AGENTS.md instructions for foxzool/openlark, covering openlark knowledge base, structure, where to look, conventions and 模块组织.
larksnap CLAUDE.md
Claude Code instructions for AmbroseX/larksnap, covering 基本约定, plan 文件管理, 开发规范, 根据场景动态调整顺序 and 文档写入.
Claude-to-IM-skill CLAUDE.md
Instructions for op7418/Claude-to-IM-skill, covering claude.md — project guidelines for claude-to-im-skill and replying to github issues.
lodestar AGENTS.md
AGENTS.md instructions for leviyuan/lodestar, covering lodestar 项目指引, 目录, 开发规则, 模块边界 and 运行中的 daemon.
cc-connect-next AGENTS.md
AGENTS.md instructions for timmyagentic/cc-connect-next, covering cc-connect development guide, project overview, architecture, key design principles and core interfaces.