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/xelandernt/skilly/agents-mdgit clone --depth 1 https://github.com/xelandernt/skillyWhat 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 | $0.01439 | $0.01439 |
| Opus 5 | $0.00720 | $0.00720 |
| Sonnet 5 | $0.00288 | $0.00288 |
| Haiku 4.5 | $0.00144 | $0.00144 |
Grade A, and why
skilly 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 2d 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 — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
1. Think Before Coding
Don't assume. Don't hide confusion. Surface tradeoffs.
Before implementing:
- Identify the affected contract first: Rust core, CLI, Python API, custom filesystem protocol, or external SkillsMP/GitHub boundary.
- Treat the Agent Skills specification
as authoritative for
SKILL.mdfields, names, resources, and validation rules. - State whether a public API change is breaking. Preserve compatibility unless the user explicitly allows breaking changes; when breaking changes are allowed, remove the old surface instead of maintaining parallel APIs.
- Confirm whether a CLI path must support both interactive terminals and non-interactive automation. Never assume a TTY is available.
- For filesystem work, reason about native and custom filesystems, path traversal, partial writes, stale resources, and rollback before coding.
- If behavior could live in Rust or Python, default to Rust core logic with a thin Python binding. Explain any exception.
- If multiple interpretations remain, present them and ask rather than choosing silently.
2. Simplicity First
Minimum code that solves the problem. Nothing speculative.
- Implement domain behavior once in
src/core.rs. Native filesystem entry points and Python/custom-filesystem entry points should delegate to the same generic implementation. - Keep
src/lib.rsas a binding layer,src/skilly/_bridge.pyas a typed adapter, andsrc/skilly/repository.pyas orchestration. Do not duplicate parsing, matching, scanning, installation, or update logic across layers. - Prefer one obvious public interface. Do not add response wrappers, aliases, convenience utilities, or configuration paths that duplicate an existing capability.
- Use
SkillRepositoryfor stateful workflows and focused discovery functions for stateless reads. - Keep CLI flows small: shared destination resolution, shared core operations, explicit TTY/non-TTY branches, and no terminal UI when plain output is enough.
- Do not add abstractions for a single use or options without a demonstrated caller.
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.
- 2d ago First seen · 142 lines · 1,439 tokens per session scan A 9bd296d2ae5f
skilly AGENTS.md is an instructions file published in the GitHub repository xelandernt/skilly (2 stars, last pushed 22d ago), licensed MIT. It adds 1,439 tokens to every session, about $0.0072 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
agents AGENTS.md
Instructions for astronomer/agents, covering claude code plugin development, plugin structure, installing the plugin, add the marketplace (from repo root) and install the plugin.
agentregistry copilot-instructions.md
Instructions for agentregistry-dev/agentregistry, covering github copilot instructions and keeping the pr description up-to-date.
loom CLAUDE.md
Instructions for cosmix/loom, covering claude.md, project overview, build commands, architecture and source layout (loom/src/).
copilot AGENTS.md
Instructions for navikt/copilot, covering agents.md for navikt/copilot, what this repo is, efficiency rule, standard commands and conventions.
autovault CLAUDE.md
Instructions for autoworks-ai/autovault, covering claude.md, project, common commands, architecture and request flow.
skills AGENTS.md
Instructions for tomimor/skills, covering agents.md, repo structure, adding a new own skill, own skill conventions and adding a new vendor skill set.