Borrowing it
Nothing to install: this file belongs to saubakirov/trace-first-starter. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/saubakirov/trace-first-starter/master/CLAUDE.mdgit clone --depth 1 https://github.com/saubakirov/trace-first-starterWrote 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/saubakirov/trace-first-starter/claude-md)<a href="https://agentmods.dev/instructions/saubakirov/trace-first-starter/claude-md"><img src="https://agentmods.dev/badge/instructions/saubakirov/trace-first-starter/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/saubakirov/trace-first-starter/claude-md"><img src="https://agentmods.dev/badge/instructions/saubakirov/trace-first-starter/claude-md.svg" alt="Reviewed on agentmods" width="80" 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 | $0.01540 | $0.01540 |
| Opus 5.5 | $0.00616 | $0.00616 |
| Sonnet 5.5 | $0.00308 | $0.00308 |
| Haiku 4.5 | $0.00154 | $0.00154 |
Grade A, and why
trace-first-starter CLAUDE.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 8d 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 — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TFW — Claude Code Adapter
TFW (Trace-First Workflow)
This project follows Trace-First Workflow. The tool-agnostic core lives in .tfw/.
Version: see .tfw/VERSION.
Working material and comments. Keep raw output, logs, exports, screenshots and scratch
scripts in the system temporary directory under tfw/<task or record ID>/, never in the project,
and remove them when your work ends. A comment exists only when it carries value for its file's
reader or a program reads it, never as a note to agents, deferred work, history or an excuse
(conventions.md → Working material, Comments).
Command Routing
Root instructions are already active; do not reload them. For /tfw-*, open the matching
.claude/commands/tfw-*.md copy and its canonical workflow completely. The workflow's Read
Contract selects task state, addressed shared ranges, templates, and knowledge inputs.
At new-task Plan entry, after identifying the request and active platform and completing the
workflow's task-control and shared-rule reads, read exactly
.tfw/adapters/claude-code/coordinator.md for the owner-facing startup card and initial mode choice.
At Plan Step 5 validate that choice and current capability; re-read this selected profile only
when the active surface or relevant capability materially changes. The task Coordinator uses the selected profile for new work. Do not preload other profiles or the tooling
manifest, and do not load this profile for every role command. A missing or ambiguous selected
pointer refuses a provider-specific offer and is reported without guessing.
For new work, one task Coordinator handles a one-phase task; long work uses distinct phase
Coordinators with bounded upward returns. Historical GATEWAY/LEAD carriers remain readable,
but a principal, title or host never supplies a route or mandate.
Activation and routing
At new-task entry, report provision · addressed send · wait/readback · title/readback from the
mechanisms exposed in the current Claude Code session, classifying each as native,
owner-assisted, or unavailable. Starting one session does not provision distinct addressable TFW
role units. Without an exposed distinct-unit creator and exact recipient send, provision and
addressed send are owner-assisted; resume/agent or title options count as native only when the
current surface also exposes their exact readback. Never borrow Codex or receipt capabilities.
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.
- 8d ago Changed · +6 lines · +87 tokens per session c3619254ccc0
- 14d ago Changed · +6 lines · +98 tokens per session 3c19b6a34f3f
- 18d ago Changed · +1 lines · +31 tokens per session 8c80f19f5e85
- 22d ago Changed · +2 lines · +43 tokens per session 1db21d709cb6
- 27d ago Changed · +12 lines · +206 tokens per session 79dbea236212
- 28d ago Changed · +19 lines · +346 tokens per session 792e421ad4ea
- 1mo ago Changed · -1 lines · -31 tokens per session e3be92f93d03
- 1mo ago Changed · -3 lines · -41 tokens per session 33f8aaf00a96
- 1mo ago Changed · -1 lines · -31 tokens per session 2c77f63c7ee9
- 1mo ago First seen · 56 lines · 832 tokens per session scan A 476ac08c90aa
trace-first-starter CLAUDE.md is an instructions file published in the GitHub repository saubakirov/trace-first-starter (30 stars, last pushed 9d ago), licensed MIT. It adds 1,540 tokens to every session, about $0.0062 per session on Opus 5.5. 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 2026-08-30.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.