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/apollographql/skills/docsgit clone --depth 1 https://github.com/apollographql/skillsWhat 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 | $0.08847 | $0.08847 |
| Opus 5 | $0.04424 | $0.04424 |
| Sonnet 5 | $0.01769 | $0.01769 |
| Haiku 4.5 | $0.00885 | $0.00885 |
Grade A, and why
skills docs.instructions.md scanned grade A with 1 finding 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- `curl`: Not "cURL". Don't use as a verb. Instead say "run the following curl command". How it starts
The opening of the file, as written. The whole thing — 703 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Preamble
The rules in this document apply to all "documentation-style" content. That can be markdown files (.md or .mdx), JSDoc or TSDoc docblocks (/** ... */), Rust DocComments /// ..., Python docstrings (""" ... """), or any other similar format.
They do not need to be applied to code or to comments that are not intended for end users (e.g., comments in source code meant for developers working on the codebase). Examples of this would be JavaScript inline comments (//), normal JavaScript block comments (/* ... */), Python comments (# ...), or Rust comments (// or /* ... */).
Primary Directive
Generate documentation content strictly adhering to the Apollo Voice and all rules in this guide.
Voice
The Apollo voice is:
- Approachable
- Positive
- Encouraging
- Helpful
- Opinionated/Authoritative
Opinionated voice prescribes a specific "happy path" to accomplish a goal.
Do this: "To achieve optimal performance, configure your server with X." Do this: "The recommended approach is to use Apollo Federation for scaling." Do this: "Always secure your GraphQL endpoint using authentication." Don't do this: "You can configure your server with X, Y, or Z." Reason: This is unopinionated and lays out options rather than prescribing a path. Don't do this: "There are several ways to scale, including Federation or stitching." Reason: This is unopinionated and presents multiple options without guiding the user. Don't do this: "Consider securing your GraphQL endpoint; options include authentication and authorization." Reason: This is unopinionated and doesn't prescribe a specific security measure.
The Apollo voice is not:
- Sarcastic
- Condescending
- Robotic
- Exaggerated
- Unopinionated
Guiding Principle: Communicate as if with a coworker you haven't met yet.
Language
- Spelling and Grammar: Use American English. Do this: "The request is canceled." Do this: "The client is customizing their schema." Do this: "The solution centralizes data management." Don't do this: "The request is cancelled." Reason: Uses British English spelling. Don't do this: "The client is customising their schema." Reason: Uses British English spelling. Don't do this: "The solution centralises data management." Reason: Uses British English spelling.
- Simplicity: Keep language simple and avoid idioms. Do this: "The client's query was very fast." Do this: "This feature makes development easier." Do this: "The system is now fully functional." Don't do this: "The client's query was lightning fast." Reason: Uses an idiom that may not be understood by non-native English speakers. Don't do this: "This feature is a piece of cake for developers." Reason: Uses an idiom. Don't do this: "The system is now firing on all cylinders." Reason: Uses an idiom.
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 · 703 lines · 8,847 tokens per session scan A 51e0b5540811
skills docs.instructions.md is an instructions file published in the GitHub repository apollographql/skills (108 stars, last pushed 1mo ago), licensed MIT. It adds 8,847 tokens to every session, about $0.0442 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
sdks AGENTS.md
AGENTS.md instructions for morpho-org/sdks, covering engineering rules, 1. architecture, layering: one direction, no exceptions, modularity: one reason to exist and testability: pure cores, i/o at the edge.
sdks CLAUDE.md
Claude Code instructions for morpho-org/sdks, a project described as: Collection of SDKs to ease interactions with the Morpho protocol and Morpho Vaults.
payload CLAUDE.md
Instructions for payloadcms/payload, covering claude.md, project structure, key directories, architecture notes and quick start.
pentagi CLAUDE.md
Instructions for vxcontrol/pentagi, covering claude.md, core interaction rules, project overview, build & development commands and backend (run from backend/).
payload AGENTS.md
Instructions for payloadcms/payload, a project described as: Payload is the open-source, fullstack Next.js framework, giving you instant backend superpowers. Get a full TypeScript backend and admin panel instantly. Use Payload as a headless CMS or for building powerful applications.
NornicDB AGENTS.md
Instructions for orneryd/NornicDB, covering nornicdb development guide for ai agents, core philosophy, 🎯 golden rules, 1. prove value before merging and 2. test-driven bug fixes.