apifable AGENTS.md

A set of repository instructions for apifable, including required versions, formatting rules, build and test commands, configuration precedence, and project-specific constraints.

In plain words
What is it for?
Use it when working on apifable to choose the right Node.js, Zod, MCP, bundler, formatting, linting, type-checking, and test conventions.
Why use it?
It gives coding agents the rules they must follow in that repository and records behaviors that are easy to miss during changes.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/ycs77/apifable/agents-md
Clone the repo
git clone --depth 1 https://github.com/ycs77/apifable

Made for: Codex, OpenCode.

Per session 771 This file is loaded in full into every session.
When invoked 771 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00771 $0.00771
Opus 5 $0.00385 $0.00385
Sonnet 5 $0.00154 $0.00154
Haiku 4.5 $0.00077 $0.00077

Measured 2d ago against content hash 36da8bd3c774, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

apifable 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 2d 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.

AGENTS.md · 49 lines

How it starts

The opening of the file, as written. The whole thing — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

Keep This File Strict

  • Do not add repo tours, architecture summaries, or file inventories.
  • Only keep constraints, non-obvious behavior, and things an agent must not do.
  • If a fact is easy to find with grep, it probably does not belong here.

Constraints

  • Node.js v22+
  • Zod v4
  • MCP SDK protocol version 2025-11-25 (stable)
  • Bundler is tsdown, not tsc
  • TypeScript formatting: 2 spaces, single quotes, no semicolons, trailing commas

Commands

  • pnpm build
  • pnpm lint --fix
  • pnpm lint [...files] --fix
  • pnpm fmt
  • pnpm fmt [...files]
  • pnpm typecheck
  • pnpm test --run

Rules

  • Config priority: CLI flag > config > default value
  • Prefer pure-function tests first; controlled temp-dir fs tests are allowed, but skip MCP handlers, CLI code, prompt flows, process.exit paths, and real network calls
  • Do not reintroduce ParsedSpec cache assumptions; parsed specs are built directly

Non-Obvious Behavior

  • HTTP_METHODS and isValidHttpMethod live in src/http-methods.ts; do not re-declare method lists in tools or parsers
  • search_endpoints default limit is 10; max is 100
  • search_schemas default limit is 10; max is 100
  • search_endpoints and search_schemas return real match count in total; truncated responses use hasMore
  • search_endpoints fuzzy fallback uses MiniSearch with fuzzy: 0.2 and prefix matching; response includes matchType, fuzzy results include score, empty fuzzy results include message
  • get_endpoint validates mutually exclusive modes in the tool function, not the handler; valid modes are method + path or operationId; empty operationId is an error
  • get_endpoint, get_types, get_schema, and list_endpoints_by_tag use src/tools/suggestions.ts for guided not-found errors; findSimilarNames tries substring matches before MiniSearch fuzzy fallback and returns up to 5 suggestions
  • list_endpoints_by_tag supports limit and offset; response includes total, offset, hasMore; warn when total > 30 && limit === undefined
  • get_types accepts exactly one mode per call: schemas, method + path, or operationId; mixing modes is an error
  • get_types returns self-contained TypeScript declarations as code text with no imports
  • get_types supports inline JSON request/response schemas through src/tools/inline-schemas.ts; names are {PascalOperationId}Request|Response or {Method}{NormalizedPath}Request|Response when there is no operationId
  • Inline schema extraction only promotes top-level non-$ref JSON request schemas and first 2xx response schemas to inline roots; nested $ref values still use normal dependency tracking
  • Zod schemas use .loose() to allow unknown keys; path-specific errors for invalid known field types (not .passthrough() which is deprecated in Zod v4)
  • Auth headers live in .apifable/auth.json and are not version-controlled; spec.headers in config is for non-secret headers; auth values override config on key conflict; both support ${ENV_VAR} expansion and keep undefined vars unchanged

Read the full file on GitHub · 49 lines

Changes

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.

  1. 2d ago First seen · 49 lines · 771 tokens per session scan A 36da8bd3c774

Subscribe to this mod's changes

apifable AGENTS.md is an instructions file published in the GitHub repository ycs77/apifable (10 stars, last pushed 16d ago), licensed MIT. It adds 771 tokens to every session, about $0.0039 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.