Borrowing it
Nothing to install: this file belongs to bcanfield/agentic-tech-debt. 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/bcanfield/agentic-tech-debt/main/CLAUDE.mdgit clone --depth 1 https://github.com/bcanfield/agentic-tech-debtWrote 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/bcanfield/agentic-tech-debt/claude-md)<a href="https://agentmods.dev/instructions/bcanfield/agentic-tech-debt/claude-md"><img src="https://agentmods.dev/badge/instructions/bcanfield/agentic-tech-debt/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/bcanfield/agentic-tech-debt/claude-md"><img src="https://agentmods.dev/badge/instructions/bcanfield/agentic-tech-debt/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 | $0.03298 | $0.03298 |
| Opus 5.5 | $0.01319 | $0.01319 |
| Sonnet 5.5 | $0.00660 | $0.00660 |
| Haiku 4.5 | $0.00330 | $0.00330 |
Grade B, and why
agentic-tech-debt CLAUDE.md scanned grade B with 2 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 1mo 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
- **Claude Code:** `tail -f ~/.claude/plugins/data/debt-ops-agentic-tech-debt/cache/<repo-hash>/debug.log` Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
- **Python over Bash for plugin scripts.** Stdlib `json`, `re`, and `subprocess.run(..., timeout=...)` beat hand-rolled JSON escaping, a `jq` dependency, and the BSD/GNU `timeout` portability dance. And run on Windows wi How it starts
The opening of the file, as written. The whole thing — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project Rules
- Stay in scope. Do exactly what was asked. No drive-by refactors or "while I'm here" cleanups. Flag them and wait for confirmation.
- Don't over-engineer. Simplest thing that works. No speculative abstractions, defensive code for impossible cases, or new dependencies without a concrete reason.
- Use current docs. For any library, framework, or API, fetch docs via the context7 MCP server before answering. Even for things you "know." Match the version in this project, not the latest release. Also - try to fetch examples from github frequently (e.g. an Anthropic-published Claude Code plugin, or an OpenAI Codex plugin, to use as an example).
- Confront our research. Try to see if you can ground decisions in our extensive research.
- Ask when unsure. One focused question beats guessing or expanding scope.
- Be refreshingly concise. Nobody likes overly wordy AI slop. Speak and comment like a co-worker.
- Guide the agent with markdown, not orchestration. Claude Code and Codex are smart and have the keys to the user's repo. Skills should load research-backed principles and trust the agent's judgment, not script its every step. Reach for Python only when determinism demands it (timestamps, audits, atomic file ops, hook contracts), never to make decisions a smart agent reading the repo can make better.
- Python over Bash for plugin scripts. Stdlib
json,re, andsubprocess.run(..., timeout=...)beat hand-rolled JSON escaping, ajqdependency, and the BSD/GNUtimeoutportability dance. And run on Windows without WSL. - Conventional commits. All four adapters' plugin versions auto-bump in lockstep via release-please (config in
.github/; mirrorsclaude-code/.claude-plugin/plugin.json,codex/.codex-plugin/plugin.json,copilot/plugin.json, andcursor/.cursor-plugin/plugin.json); usefeat:/fix:/feat!:prefixes. Anything else is ignored by the bumper. Don't hand-editversionin anyplugin.json. - We like humanlike, concise inline comments For example, a single line above a function or code block very simply and concisely saying what the code is doing so that the code is easy for a human to understand
- Record decisions as ADRs. When we make a choice with two credible alternatives (a new convention, a tradeoff worth remembering), drop a short note in
docs/adr/using the format in its README.
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.
- 1mo ago First seen · 151 lines · 3,298 tokens per session scan B 76706fa5a0ff
agentic-tech-debt CLAUDE.md is an instructions file published in the GitHub repository bcanfield/agentic-tech-debt (8 stars, last pushed 2d ago), licensed MIT. It adds 3,298 tokens to every session, about $0.0132 per session on Opus 5.5. A static security scan graded it B with 2 findings (reads agent configuration directories, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
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.
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.
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).