Borrowing it
Nothing to install: this file belongs to abhiyoheswaran1/tokentrace. 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/abhiyoheswaran1/tokentrace/main/AGENTS.mdgit clone --depth 1 https://github.com/abhiyoheswaran1/tokentraceWrote 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/abhiyoheswaran1/tokentrace/agents-md)<a href="https://agentmods.dev/instructions/abhiyoheswaran1/tokentrace/agents-md"><img src="https://agentmods.dev/badge/instructions/abhiyoheswaran1/tokentrace/agents-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/abhiyoheswaran1/tokentrace/agents-md"><img src="https://agentmods.dev/badge/instructions/abhiyoheswaran1/tokentrace/agents-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 | $0.01004 | $0.01004 |
| Opus 5 | $0.00502 | $0.00502 |
| Sonnet 5 | $0.00201 | $0.00201 |
| Haiku 4.5 | $0.00100 | $0.00100 |
Grade A, and why
tokentrace 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 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 — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TokenTrace Agent Instructions
These instructions apply to coding agents working in this repository.
Mandatory Workflow
- Use the Superpowers methodology from https://github.com/obra/superpowers for coding work.
- Before non-trivial implementation, use the relevant Superpowers skills for brainstorming, systematic debugging, test-driven development, planning, code review, and verification.
- Prefer TDD for behavior changes: write a failing test, verify it fails for the expected reason, implement the smallest fix, and verify it passes.
- Use ProjScan from https://www.npmjs.com/package/projscan as a standard code-intelligence and quality gate.
- Run targeted
projscancommands when exploring substantial changes, and runnpm run projscan:doctorafter substantial implementation or before release readiness claims. - ProjScan does not replace tests, typecheck, lint, build, package, or smoke verification.
Release Discipline
- Do not bump
package.json, updatepackage-lock.jsonfor a version bump, create git tags, push release tags, create GitHub Releases, or publish to npm unless the maintainer explicitly asks for a release. - Build and test substantial improvements before any release conversation.
- User-facing changes must be recorded in
CHANGELOG.mdunderUnreleasedwhile in development. - Read
docs/RELEASE_CHECKLIST.mdbefore any release work. - Final public release gates include
npm run release:check, packed-install smoke when needed, ProjScan doctor, changelog section extraction, tag/version matching, GitHub Trusted Publishing, and npm verification.
Maintenance Pass
TokenTrace is in maintenance mode: feature-complete, iterate from feedback. New features are pull-based (wait for real usage signals); health is push-based (run proactively). A maintenance pass is read-only triage — it reports and proposes, it does not bump versions or release. Run it periodically or when asked for a "health check".
Steps:
- Test/type/lint gate:
npm run verifymust be fully green (vitest +tsc --noEmit+ eslint). Treat any failure as a regression to fix before other work. - Dependency drift:
npm outdated. Patch/minor bumps within the current major are low-risk; batch and verify them. Hold major bumps (e.g. eslint 10,@vitejs/plugin-react6,@types/node25,eslint-plugin-react-hooks7) for a deliberate, separately-tested upgrade — never fold a major into a routine pass. - Security:
npm audit --audit-level=moderate(expect 0 vulnerabilities) andnpm run security:ioc. For release-grade checks usenpm run security:package. - Provider/ecosystem drift (highest-value signal for a cost tool): check
pricing/default-model-prices.jsonfreshness and refresh withnpm run pricing:refreshwhen models, tokenizers, or provider pricing have changed upstream. This is "feedback from reality," not from users — act on it without waiting for a complaint. - Package surface:
npm pack --dry-runand confirm the tarball contains only intended files. Nothing untracked-in-git should reach npm. Thefilesarray includespublic/, so anything dropped intopublic/ships — keep non-product assets (brand kits, screenshots, scratch files) out of it or.gitignorethem. - ProjScan:
npm run projscan:doctorfor code-intelligence and release-readiness signal. - Report, don't release: summarize findings and propose fixes. Apply only low-risk, verified maintenance (patch bumps, lockfile, packaging hygiene) under the Release Discipline rules above; leave version bumps and publishing to an explicit maintainer request.
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 First seen · 49 lines · 1,004 tokens per session scan A 9b5cdd3a2abe
tokentrace AGENTS.md is an instructions file published in the GitHub repository abhiyoheswaran1/tokentrace (0 stars, last pushed 2mo ago), licensed MIT. It adds 1,004 tokens to every session, about $0.0050 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
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.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
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).
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
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.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.