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/masihmoafi/elpis/agents-mdgit clone --depth 1 https://github.com/MasihMoafi/ElpisWrote 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/masihmoafi/elpis/agents-md)<a href="https://agentmods.dev/instructions/masihmoafi/elpis/agents-md"><img src="https://agentmods.dev/badge/instructions/masihmoafi/elpis/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 | $0.01007 | $0.01007 |
| Opus 5 | $0.00504 | $0.00504 |
| Sonnet 5 | $0.00201 | $0.00201 |
| Haiku 4.5 | $0.00101 | $0.00101 |
Grade A, and why
Elpis 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 — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Elpis Agent Map
Start
- Read
docs/GUIDE.md; it is the product, architecture, and requirements source of truth. - Read
TASKS.md; work only on its Current Action unless Masih changes priority. - Read
docs/context.mdanddocs/sessions.mdbefore touching context, session, or pruning behavior. - Read
docs/SHIPPING_RULES.mdbefore a release or any change that reads the filesystem, environment, or network. - Read
docs/LOCAL_BUILD_RULES.mdbefore runningcargoon Masih's workstation. - Verify repository state before editing and preserve unrelated user changes.
- Challenge unclear or solution-first requirements with
$challenge-requirementsbefore planning implementation.
Context Discipline
- Load only the guide sections and source files needed for the current task.
- Keep the active goal, changed files, verification, blocker, and next action visible.
- Summarize terminal output; do not carry raw logs once their result is known.
- After edits, retain the diff and verification result; reread file bodies on demand.
- Do not add slash commands unless Masih explicitly selects them.
- Worker agents must not edit
GUIDE.mdorTASKS.md; the coordinator owns those files. - Do not delegate to Jules. The coordinator selects and manages workers and worktrees.
Definition of Done
- Masih is the sole arbiter of truth. CI, compilation, tests, and agent claims are evidence, not functional acceptance.
- Before important or difficult implementation work, turn the request into an explicit acceptance harness and confirm it with Masih.
- Implement, run the required automated checks, build/install when applicable, and hand Masih a plain user test checklist.
- A task becomes verified only after Masih accepts the user-visible behavior and the evidence is recorded in
TASKS.md. - Behavior must be implemented, not merely documented.
- Rust changes pass the applicable Rust checks. This repository contains no Python.
- Known gaps and skipped checks are stated plainly.
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 · 95 lines · 1,007 tokens per session scan A 028d07bdf079
Elpis AGENTS.md is an instructions file published in the GitHub repository MasihMoafi/Elpis (41 stars, last pushed 3d ago), licensed Apache-2.0. It adds 1,007 tokens to every session, about $0.0050 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
ollama CLAUDE.md
Claude Code instructions for ollama/ollama: See AGENTS.md for the shared agent instructions for this repository.
pocketpaw CLAUDE.md
Instructions for pocketpaw/pocketpaw, covering claude.md, project overview, knowledge base, search the kb from terminal and show a specific module's wiki.
pocketpaw copilot-instructions.md
Instructions for pocketpaw/pocketpaw, covering github copilot instructions for pocketpaw, 1. event-driven message bus, 2. async everywhere, ✅ correct and ❌ wrong — blocks the event loop.
codexia AGENTS.md
Instructions for milisp/codexia, covering agents.md, project info, project tech, common commands and project structure.
ai4j AGENTS.md
Instructions for LnYo-Cly/ai4j, covering repository guidelines, project identity, monorepo scope, harness anything and hard rules.
cognithor CLAUDE.md
Instructions for Alex8791-cyber/cognithor, covering claude.md, what this is, source layout, architecture in one paragraph and critical conventions.