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/nanookai/worldbank-api/agents-mdgit clone --depth 1 https://github.com/NanookAI/worldbank-apiWrote 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/nanookai/worldbank-api/agents-md)<a href="https://agentmods.dev/instructions/nanookai/worldbank-api/agents-md"><img src="https://agentmods.dev/badge/instructions/nanookai/worldbank-api/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 | $0.01106 | $0.01106 |
| Opus 5 | $0.00553 | $0.00553 |
| Sonnet 5 | $0.00221 | $0.00221 |
| Haiku 4.5 | $0.00111 | $0.00111 |
Grade A, and why
worldbank-api AGENTS.md 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.
(`curl "https://api.worldbank.org/v2/...?format=json"`) before it goes into How it starts
The opening of the file, as written. The whole thing — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Instructions
This repository contains an Agent Skill that teaches AI agents how to use the World Bank Indicators API (https://api.worldbank.org/v2). The deliverable is documentation, not application code.
Layout
skills/worldbank-api/SKILL.md— skill entry point (frontmatter, data-query workflow, top indicator codes, error handling, tips)skills/worldbank-api/scripts/wb_data.py— zero-dependency CLI covering the common case (indicator + countries → formatted series) plus client-side indicator searchskills/worldbank-api/references/endpoints.md— full endpoint catalog: data queries, countries, indicators, topics, sources, regions, income levels, monthly/quarterly data, languages, paginationskills/worldbank-api/references/indicators.md— verified indicator codes grouped by topic
Conventions
- English only in all files. No Chinese or other non-English text in any committed file, including examples and comments.
- Verify before documenting. Every endpoint, parameter, and indicator code
must be checked against the live API
(
curl "https://api.worldbank.org/v2/...?format=json") before it goes into the docs. Do not copy indicator codes from memory — the World Bank archives series without notice (e.g. allEN.ATM.CO2E.*CO2 codes and the entire Doing Business source are gone), and an archived code fails only on the data query (error id 175), not the metadata query. - Keep SKILL.md lean (well under 500 lines). SKILL.md carries the workflow
and the ~16 most-used codes; the exhaustive endpoint catalog and the larger
code list belong in
references/, with clear pointers from SKILL.md. - Explain the why. When documenting a pitfall, state the reason (e.g. "prefer mrnev over mrv because most series lag 1–2 years") so agents can generalize instead of pattern-matching.
- Free API is the only stance. The Indicators API is keyless with no paid tier; never invent authentication steps.
- The script stays dependency-free.
wb_data.pymust run on a bare Python 3.8+ standard library (urllib, json, argparse) so it works in any agent sandbox without a pip install.
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 · 75 lines · 1,106 tokens per session scan A 0dabcf2f0155
worldbank-api AGENTS.md is an instructions file published in the GitHub repository NanookAI/worldbank-api (2 stars, last pushed 1mo ago), licensed MIT. It adds 1,106 tokens to every session, about $0.0055 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 2026-08-31.
Other instructions, from other repositories
aquascope copilot-instructions.md
Copilot instructions for Rekin226/aquascope, covering copilot instructions for aquascope, build, test, and lint, run all tests, run a single test file and run a single test by name.
openpitch AGENTS.md
AGENTS.md instructions for Avierovich/openpitch, covering agent coordination log, current coordination rules, current state (2026-06-26), codex edits log and 2026-06-13 - product/strategy documentation expansion.
swiss-public-data-mcp CLAUDE.md
Instructions for malkreide/swiss-public-data-mcp, covering project conventions for claude, how verification works here, check the result, not the activity, run the check that counts, not the one you know and the ruff pin: one source per repository.
kpubdata AGENTS.md
AGENTS.md instructions for yeongseon/kpubdata, covering agents.md, 목적, 먼저 읽을 문서, 작업 원칙 and 언어 정책.
luxembourg-mcp CLAUDE.md
Instructions for amirdaraee/luxembourg-mcp, covering claude.md, commands, hard constraint: zero dependencies, architecture and adding or changing a tool touches five places.
eurostat-mcp-server AGENTS.md
AGENTS.md instructions for cyanheads/eurostat-mcp-server, covering agent protocol, what's next?, core rules, patterns and tool.