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/robcsaszar/scaffold/agents-mdgit clone --depth 1 https://github.com/robcsaszar/scaffoldWhat 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 | $0.00525 | $0.00525 |
| Opus 5 | $0.00262 | $0.00262 |
| Sonnet 5 | $0.00105 | $0.00105 |
| Haiku 4.5 | $0.00052 | $0.00052 |
Grade A, and why
scaffold 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.
How it starts
The opening of the file, as written. The whole thing — 33 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Mission
This repo publishes the scaffold skill pack: three skills for creating and reviewing the standing context files (AGENTS.md, CONTEXT.md, DESIGN.md) an AI agent reads before working on a codebase. There is no build, no tests, no runtime of its own. The deliverable is the contents of skills/*, each copied in from a source repo (Orakl) where they are used and refined, then published here for anyone to install. Changes should be judged by: would a stranger who installs one of these skills into an unrelated project get correct, generic guidance out of it?
Layout convention
- One directory per skill under
skills/<name>/,name:in frontmatter matching the directory name exactly. - Each skill's full directory ships as-is:
SKILL.mdplus anyreferences/it depends on. Don't split a skill across a partial copy. - No skill in this pack ships a
scripts/directory. If a future update adds one, add a SAFETY.md documenting it in the same change (see howai-forge, a sibling pack, handles this).
Judgment boundaries
NEVER:
- Never let a copied skill's content leak repo-specific references back to Orakl (or any other single source project). A published skill needs to read as generic guidance, not one project's internal notes.
- Never remove a skill without asking first: this repo's whole value is the stable set of three, and removing one is a breaking change for anyone who installed it.
ASK:
- Ask before changing the license or copyright holder.
- Ask before adding a fourth skill that doesn't fit the "standing context file" shape the other three share.
ALWAYS:
- When re-syncing a skill from its source, re-run the orakl-leakage grep (
grep -rni "orakl"acrossskills/) before publishing, and flag any hits rather than silently editing them out. - When adding, removing, or renaming a skill: update the table in
README.mdin the same change.
Adding a skill
- Copy the full skill directory from its source, not just
SKILL.md. - Grep the copy for source-repo-specific leakage (project name, absolute paths) before publishing.
- Add a row to the table in
README.md. - Decide whether it ships a
scripts/directory; if so, addSAFETY.mddescribing what the script does.
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.
- yesterday First seen · 33 lines · 525 tokens per session scan A f7e21a546f2a
scaffold AGENTS.md is an instructions file published in the GitHub repository robcsaszar/scaffold (0 stars, last pushed 3d ago), licensed MIT. It adds 525 tokens to every session, about $0.0026 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
Skill_Seekers AGENTS.md
Instructions for yusufkaraaslan/Skill_Seekers, covering agents.md - skill seekers, project overview, key capabilities, setup and with dev tools (pytest, ruff, mypy, coverage).
aspens copilot-instructions.md
Instructions for aspenkit/aspens, covering copilot instructions, what this project does, project setup, architecture and doc init pipeline.
agent-stylebooks AGENTS.md
Instructions for Neeeophytee/agent-stylebooks, a project described as: 11 installable editorial systems for AI agents, based on leading public style guides.
docs-to-mcp AGENTS.md
AGENTS.md instructions for gabrielassisxyz/docs-to-mcp, covering agents.md — read this first, build / run / test, conventions and common hurdles (append as discovered).
agentic-project-template AGENTS.md
Instructions for pueraeternis/agentic-project-template, covering agents.md, ai agent guide, purpose, repository philosophy and framework overview.
pituitary AGENTS.md
AGENTS.md instructions for nantobv/pituitary, covering agents.md, project, org-wide policy (nantobv-shared), coding standards and architecture rules.