agentmods.dev

Throughline AGENTS.md

A set of instructions for agents working in the Throughline repository. Throughline is a standalone tool that handles installation, saved state, database schema changes, diagnosis, recovery, updates, and release decisions.

In plain words
What is it for?
Use it before modifying Throughline source code, tests, hooks, commands, database handoffs, or support for other coding hosts. It also points to the relevant design documents for specialized work.
Why use it?
It tells agents which files and documents to read first and which parts of the product Throughline owns. This prevents changes that replace or bypass its existing Claude Code behavior and data contracts.

Instructions file for CodexOpenCode

2 2d ago A scan 2,267 tokens original MIT
View the source file kitepon/Throughline
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/kitepon/throughline/agents-md
Clone the repo
git clone --depth 1 https://github.com/kitepon/Throughline

Made for: Codex, OpenCode.

Per session 2,267 This file is loaded in full into every session.
When invoked 2,267 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 $0.02267 $0.02267
Opus 5 $0.01133 $0.01133
Sonnet 5 $0.00453 $0.00453
Haiku 4.5 $0.00227 $0.00227

Measured yesterday against content hash 2b63e2f991b8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

Throughline 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 yesterday.

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 · 117 lines

How it starts

The opening of the file, as written. The whole thing — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

このファイルは Codex など、Claude Code 以外のエージェントがこのリポジトリで作業する際の入口です。

正本は CLAUDE.md です。

Throughline は Claude Code を主軸として育ってきたプロジェクトです。Codex 対応や rollback-based trimming を進める場合も、Claude 向けの hook、slash command、transcript contract、handoff behavior を Codex 用に置き換えてはいけません。

最初に読むもの

作業前に、最低限次を読むこと。

  1. CLAUDE.md と docs/00_overview.md
  2. 変更対象の source / test
  3. Codex 両対応や context trim に関わる作業なら:
  4. DB記憶を別プロセスへ渡す作業なら:
  5. Grok host / /tl 後継なら:
  6. Cursor host なら:

CLAUDE.md と実装が食い違う場合は、まず実装を直接確認する。必要なら CLAUDE.md 側を更新する。

完了済みの計画・実装記録は docs/archive/ に置く。通常作業では読まず、 設計判断の経緯や過去の受入証拠が必要なときだけ参照する。

製品の所有境界

Throughline は単独で install、設定、状態保存と schema migration、診断、復旧、更新、 release 判定まで完結する。正規入口は本repositoryの README、CLI、source、release手順である。 dotagents は工場全体の統合契約と配線を担うが、Throughline の状態やschemaを所有・制御せず、 単独運用の前提にもならない。外部連携は公開CLI・JSON契約だけを使い、SQLiteを直接操作しない。

Grok は first-class host である。capture は chat_history.jsonl、hook は ~/.grok/hooks/throughline.json の絶対パス。Grok /tl の記憶再開は throughline grok-continue --session grok:<id> だけとする。stdout 再注入、 aiterm、--rules、--from を現行契約に戻さない。

Cursor は first-class host である。capture は agent-transcripts jsonl、hook は ~/.cursor/hooks.json への upsert（工場 hook は残す）。絶対 node + bin/throughline.mjs。引き継ぎ注入は sessionStart の additional_context。 /tl 後継の自動起動はしない。

Claude 正本を守る

  • .claude/、Claude hooks、/tl、/sc-detail、Claude transcript parsing、Claude handoff / resume behavior は first-class のまま維持する。
  • Codex 対応のために Claude-facing field、command name、hook shape、session / baton semantics を rename しない。
  • 既存 Claude path を削除、劣化、Codex path への暗黙置換をしない。
  • Codex 向け表現が必要な場合は、既存データを変更するのではなく adapter / projection を追加する。
  • Claude contract に触れる変更では、先に現在挙動を固定する test / fixture を置く。

Read the full file on GitHub · 117 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. yesterday First seen · 117 lines · 2,267 tokens per session scan A 2b63e2f991b8

Subscribe to this mod's changes

Throughline AGENTS.md is an instructions file published in the GitHub repository kitepon/Throughline (2 stars, last pushed 2d ago), licensed MIT. It adds 2,267 tokens to every session, about $0.0113 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.

DebuggingGit and workflowTestingai-codingclaudeclaude-codeclaude-code-hookscontext-compressionhooks
Related

Other instructions, from other repositories

dev-flow AGENTS.md

Instructions for daphnee-ovo/dev-flow, covering dev-flow plugin, language policy, 注意事项, 命令 and 流程.

daphnee-ovo/dev-flow · 1,675 tokens

cc-safe-setup CLAUDE.md

Instructions for yurukusa/cc-safe-setup, covering project rules, safety, code style and git.

yurukusa/cc-safe-setup · 98 tokens

claude_extensions CLAUDE.md

Instructions for kimon1230/claude_extensions, covering global claude.md, persona, hard rules, code quality and planning.

kimon1230/claude_extensions · 1,317 tokens

UnseveredMemory CLAUDE.md

Claude Code instructions for blas0/UnseveredMemory, covering project instructions, tech stack, architecture, development and setup.

blas0/UnseveredMemory · 135 tokens

claude-hangar CLAUDE.md

Claude Code instructions for claude-hangar/claude-hangar, covering claude hangar — project instructions, what is this?, repository structure, language & communication and quality standards.

claude-hangar/claude-hangar · 1,039 tokens

skill-olympus AGENTS.md

AGENTS.md instructions for Dannykkh/skill-olympus, covering agents.md, quick retrieval paths, recommended workflows, core rules (always apply) and native-first 구현 경계 (항상 적용).

Dannykkh/skill-olympus · 7,187 tokens