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/thefrontside/effection/agents-mdgit clone --depth 1 https://github.com/thefrontside/effectionWhat 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 | $0.03630 | $0.03630 |
| Opus 5 | $0.01815 | $0.01815 |
| Sonnet 5 | $0.00726 | $0.00726 |
| Haiku 4.5 | $0.00363 | $0.00363 |
Grade A, and why
effection 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 — 458 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — Effection agent contract
This file is the behavioral contract for AI agents working with the Effection codebase.
Agents must not invent APIs, must not infer semantics from other ecosystems, and must ground claims in the public API and repository code.
If you are unsure whether something exists, consult the API reference: https://frontside.com/effection/api/
Core invariants (do not violate)
Operations vs Promises
- Operations are lazy. They execute only when interpreted (e.g.
yield*,run(),Scope.run(),spawn()). - Promises are eager. Creating a promise (or calling an
asyncfunction) starts work;awaitonly observes completion. - You must not claim that a promise is "inert until awaited".
- You must not use
awaitinside a generator function (function*). Useyield*with an operation instead (e.g.yield* until(promise)).
Structured concurrency is scope-owned
- Scope hierarchy is created automatically by the interpreter; application code should not manage scopes manually.
- "Lexical" in Effection: scope hierarchy follows the lexical structure of
operation invocation sites (e.g.
yield*,spawn,Scope.run), not where references are stored or later used. - Work is owned by Scopes.
- When a scope exits, all work created in that scope is halted.
- References do not extend lifetimes. Returning a
Task,Scope,Stream, orAbortSignaldoes not keep it alive.
Effects do not escape scopes
- Values may escape scopes.
- Ongoing effects must not escape: tasks, resources, streams/subscriptions, and context mutations must remain scope-bound.
Operations, Futures, Tasks
Operation
- An
Operation<T>is a recipe for work. It does nothing by itself. - Operations are typically created by invoking a generator function
(
function*).
Future
- A
Future<T>is both:- an Effection operation (
yield* future) - a Promise (
await future)
- an Effection operation (
Task
- A
Task<T>is aFuture<T>representing a concurrently running operation. - A task does not own lifetime or context; its scope 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.
- 2d ago First seen · 458 lines · 3,630 tokens per session scan A 8500360c3944
effection AGENTS.md is an instructions file published in the GitHub repository thefrontside/effection (848 stars, last pushed 25d ago), licensed MIT. It adds 3,630 tokens to every session, about $0.0181 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 2026-08-30.
Other instructions, from other repositories
asupersync CLAUDE.md
Claude Code instructions for Dicklesworthstone/asupersync, covering claude.md — asupersync, primary directive, quick reference and full guidelines.
asupersync AGENTS.md
AGENTS.md instructions for Dicklesworthstone/asupersync, covering agents.md — asupersync, rule 0 - the fundamental override prerogative, rule number 1: no file deletion, git branch: only use main, never master and rule 2: no git branches. no git worktrees. ever.
vera CLAUDE.md
Instructions for aallan/vera, covering claude.md — project orientation for claude code, virtual environment, key commands, project layout and writing vera code.
vera AGENTS.md
Instructions for aallan/vera, covering agents.md — instructions for ai agents, for agents writing vera code, conformance programs as reference, workflow and commands.
NeoHaskell AGENTS.md
Instructions for neohaskell/NeoHaskell, covering neohaskell — agent guide, hard rule, style (mandatory), commands (pointers, not a duplicate) and work intake.
Swift-Concurrency-Agent-Skill copilot-instructions.md
Instructions for AvdLee/Swift-Concurrency-Agent-Skill, covering github copilot instructions: agent skills expert, core agent skills principles, format compliance, skill content organization and review focus areas.