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/tyler-mitchell/type-atlas/agents-mdgit clone --depth 1 https://github.com/tyler-mitchell/type-atlasWrote 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/tyler-mitchell/type-atlas/agents-md)<a href="https://agentmods.dev/instructions/tyler-mitchell/type-atlas/agents-md"><img src="https://agentmods.dev/badge/instructions/tyler-mitchell/type-atlas/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 | $0.05987 | $0.05987 |
| Opus 5 | $0.02993 | $0.02993 |
| Sonnet 5 | $0.01197 | $0.01197 |
| Haiku 4.5 | $0.00599 | $0.00599 |
Grade A, and why
type-atlas 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 — 175 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Code Guidelines:
- Scripting Language: TypeScript script files
- Programming Style: Functional Programming
- Comments: bare minimum, caveman prose. Say the thing in the fewest plain words that still carry it — short clauses, no essays, no story of how the code got here. A comment earns its line only when the code cannot state the constraint itself. This binds every comment you write or edit; it is never an activity of its own, and never a reason to touch comments the work did not already touch.
Shared Agent Workflow
- Daily branch:
main - Bumpy base branch:
release - Generated version PR:
bumpy/version-packages
The human owns the checked-out branch. Agents never create, switch, rename, delete, reset, or replace branches unless the human requests that exact operation. If another branch is checked out, continue there and report the difference.
Work and commit on the checked-out branch. Stage only task-owned files. If the
index already contains another agent’s files, commit task-owned paths only and
leave the other staged entries untouched with
git commit --only -- <task-owned paths>. Never delete .git/index.lock; wait
for the other Git operation to finish.
commit authorizes a local commit only. push authorizes the checked-out
branch and includes every unpushed commit already on it; report that complete
commit set before pushing. Consumer-visible package changes include one
maintained Bumpy bump file. Agents never create task branches or worktrees.
Bump lifecycle
Bumps are authored during change development, never reconstructed just before
release. The first consumer-visible commit for a logical change creates one
bump file through vp run release add; later commits for that same change
update the same file. An unrelated logical change gets its own bump file.
Commit the implementation, tests, generated consumer docs, and bump file
together. Follow .skills/add-change/SKILL.md for the exact command, bump
level, package attribution, and changelog text. Use patch for compatible fixes,
minor for compatible capabilities, and major for breaking public contracts.
Name only directly changed packages; Bumpy owns fixed-group and dependency
propagation. Root shared changes name every affected public package explicitly.
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 · 175 lines · 5,987 tokens per session scan A d33b149c4c3f
type-atlas AGENTS.md is an instructions file published in the GitHub repository tyler-mitchell/type-atlas (2 stars, last pushed 5d ago), licensed Apache-2.0. It adds 5,987 tokens to every session, about $0.0299 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
idesense CLAUDE.md
Instructions for vcth4nh/idesense, covering idesense, agent rules and where things live.
llm-context.py CLAUDE.md
Instructions for cyberchitta/llm-context.py, covering claude.md, working notes (gitignored) and draining the field notes.
openrouter-mcp-multimodal AGENTS.md
AGENTS.md instructions for stabgan/openrouter-mcp-multimodal, covering agent instructions, before you ship, releasing (read this before publishing), short version and version files (must all match package.json).
seekstone CLAUDE.md
Instructions for shaqmughal/seekstone, covering claude.md, what this repo is, commands, the harness itself (run after npm install) and architecture.
Plonk AGENTS.md
Instructions for ostapondo/Plonk, covering agent rules, layout, adding a module, build & verify and code style.
second-brain AGENTS.md
Instructions for LindaHaviv/second-brain, covering instructions for ai coding agents, what to read first, the rules (non-negotiable), common tasks, the sanctioned way and enforcement, not just instructions.