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/pedropaulovc/claude-code-types/agents-mdgit clone --depth 1 https://github.com/pedropaulovc/claude-code-typesWrote 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/pedropaulovc/claude-code-types/agents-md)<a href="https://agentmods.dev/instructions/pedropaulovc/claude-code-types/agents-md"><img src="https://agentmods.dev/badge/instructions/pedropaulovc/claude-code-types/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 | $0.01220 | $0.01220 |
| Opus 5 | $0.00610 | $0.00610 |
| Sonnet 5 | $0.00244 | $0.00244 |
| Haiku 4.5 | $0.00122 | $0.00122 |
Grade A, and why
claude-code-types 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 3d 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 — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
claude-code-types
Types-only npm package for Claude Code chat history JSONL files.
Project Structure
index.d.ts # All type definitions (the package payload)
scripts/scan-transcripts.ts # Coverage scanner: finds JSONL values index.d.ts doesn't model yet
scripts/lib/extract-types.ts # Parses index.d.ts (TS compiler API) — single source of truth for known values
test/smoke.local.test.ts # Smoke test against real JSONL files (KNOWN_TYPES derived from index.d.ts)
package.json # types field, no main/module
tsconfig.json # Strict TS config for validation
scripts/tsconfig.json # Strict TS config for the scanner + test
CHANGELOG.md # Keep a Changelog format
.github/workflows/ci.yml # CI: typecheck + verify docs freshness on PRs and main pushes
.github/workflows/auto-tag.yml # Auto-creates v* tag from package.json on main push
.github/workflows/publish.yml # Single job: validate → npm publish → GitHub Release, on v* tag push
Commands
npm run typecheck— Validateindex.d.tscompiles under strict modenpm run typecheck:scripts— Typecheck the scanner and test under strict modenpm run scan— Scan~/.claude/projects/and report every discriminator/union value and entry field not yet covered byindex.d.ts, with counts and samples. Exits non-zero when anything is uncovered. Run this first when refreshing types. Flags:--dir <path>,--json,--no-samples.npm test— Run smoke tests (requires~/.claude/projects/to contain JSONL files)
Refreshing types from logs
scripts/lib/extract-types.ts parses index.d.ts as the source of truth, so the scanner and smoke test automatically track whatever the union declares — never hardcode a parallel list. To refresh:
npm run scan— lists new entry types, system subtypes, attachment types, tools, models, stop reasons, permission modes, queue ops, cache-miss reasons, content-block/citation types, and undeclared top-level fields per entry type. Each new value comes with a count and a sample JSON line.- For each reported value, add it to the location named in the report (e.g. a new
Attachmentmember + interface, or a field onEntryBase). For new entry/attachment interfaces, use the sample to type the payload. - Re-run
npm run scanuntil it reports full coverage, thennpm run typecheck && npm test.
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.
- 3d ago First seen · 72 lines · 1,220 tokens per session scan A 725bf6d27f6b
claude-code-types AGENTS.md is an instructions file published in the GitHub repository pedropaulovc/claude-code-types (2 stars, last pushed 1mo ago), licensed MIT. It adds 1,220 tokens to every session, about $0.0061 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
react-seed AGENTS.md
Instructions for guokaigdg/react-seed, covering agents.md, 目录, 1. 项目概览, 2. 技术栈 and 3. 目录结构.
persian-tools typescript.instructions.md
Instructions for persian-tools/persian-tools, covering typescript development instructions, typescript best practices for persian-tools, type safety, function signatures and type definitions.
agentic-playwright enums.instructions.md
Instructions for idavidov13/agentic-playwright, covering enums, critical, file locations, instructions and phase 1: decide if the value belongs in an enum.
ui-extensions-sdk CLAUDE.md
Instructions for contentful/ui-extensions-sdk, covering claude project instructions — ui-extensions-sdk, identity & scope, working style, sharp edges and verification before claiming done.
hana-developer-cli-tool-example mcp-server-development.instructions.md
Use when creating or updating MCP (Model Context Protocol) server components. Enforces TypeScript patterns, JSON-RPC communication rules, tool/resource/prompt registration, and integration with CLI command metadata. Ensures MCP server follows protocol requirements and maintains consistency with the CLI infrastructure.
innovation-lab-examples typescript.instructions.md
Instructions for fetchai/innovation-lab-examples, a project described as: 80+ production-ready AI agent examples in Python — build autonomous agents, multi-agent systems and agentic AI with uAgents, ASI:One, MCP, A2A, LangChain, CrewAI, Gemini, Claude and OpenAI.