agentic.nvim AGENTS.md

Project instructions for agentic.nvim, a Neovim plugin that adds AI coding help through chat. They explain which guidance files and skills to read before changing code or tests.

In plain words
What is it for?
Use them when modifying the plugin, its chat interface, ACP client, or tests. They help locate nested instructions, load required skills, and check the glossary before making architectural decisions.
Why use it?
They reduce the risk of making changes without understanding the relevant part of the project. They also help the agent resolve project-specific terms and follow its testing process.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/carlos-algms/agentic.nvim/agents-md
Clone the repo
git clone --depth 1 https://github.com/carlos-algms/agentic.nvim

Made for: Codex, OpenCode.

Per session 4,974 This file is loaded in full into every session.
When invoked 4,974 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.04974 $0.04974
Opus 5 $0.02487 $0.02487
Sonnet 5 $0.00995 $0.00995
Haiku 4.5 $0.00497 $0.00497

Measured 3d ago against content hash 1240bd8e2c71, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agentic.nvim 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.

AGENTS.md · 424 lines

How it starts

The opening of the file, as written. The whole thing — 424 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agents Guide

agentic.nvim is a Neovim plugin that emulates Cursor AI IDE behavior, providing AI-driven code assistance through a chat interface.

Nested instructions

Read these before touching the matching area:

  • lua/agentic/acp/AGENTS.md - ACP client, tool calls, permissions, providers
  • lua/agentic/ui/AGENTS.md - chat UI: topology, lifecycle contracts (open/close/destroy), MessageWriter state machines, tool-call block layout, folding, auto-scroll, permission reanchor, traps
  • tests/AGENTS.md - test framework, TDD workflow, assertions, helpers

Skill driven

This project splits instructions and repetitive workflows into skills.

MANDATORY: Load the related skill before suggesting a solution, writing to lua files, or adding, updating, or creating unit tests.

Skills hold good practices, code standards, and architecture flows that may not match your training data. Load them before assuming your solution is right.

Domain glossary — lazy read

CONTEXT.md (repo root) defines overloaded terms (Session, Agent, Provider, Tool Call, Diff, etc.). Do NOT pre-read. Grep it for the ambiguous keyword first; only Read the file if the grep matches. If no match, the term is not in the glossary — proceed without loading.

Architectural decisions (ADRs) — optional read

docs/adr/ stores Architecture Decision Records. One file per subject, NNNN-short-slug.md (4-digit). "ADR 2" = 0002-*.md. Full template in docs/adr/README.md.

Do NOT pre-read. Load only when:

  • A rule in AGENTS.md is unclear, contested, or looks arbitrary.
  • You are about to rewrite a subsystem an ADR covers.
  • A reviewer asks "why didn't we do X?".

Discovery: grep docs/adr/ for the keyword, read only on match. No match = no ADR exists.

ADR NNNN cited in a nested AGENTS.md/CONTEXT.md is an implicit match, NOT an auto-load. Read it only when a trigger above fires. Citation in passing is not a trigger.

grill-with-docs override

grill-with-docs skill ships a minimal ADR template. Path + 4-digit numbering match this repo; the template does NOT. Use docs/adr/README.md verbatim. Keep the Rejected / superseded alternatives table and Changelog. Changelog rows contain only the date and change summary.

Read the full file on GitHub · 424 lines

Changes

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.

  1. 3d ago First seen · 424 lines · 4,974 tokens per session scan A 1240bd8e2c71

Subscribe to this mod's changes

agentic.nvim AGENTS.md is an instructions file published in the GitHub repository carlos-algms/agentic.nvim (609 stars, last pushed 10d ago), licensed MIT. It adds 4,974 tokens to every session, about $0.0249 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.