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/juhaku/aiwaku.nvim/contextgit clone --depth 1 https://github.com/juhaku/aiwaku.nvimWrote 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/juhaku/aiwaku.nvim/context)<a href="https://agentmods.dev/instructions/juhaku/aiwaku.nvim/context"><img src="https://agentmods.dev/badge/instructions/juhaku/aiwaku.nvim/context.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 | $0.02247 | $0.02247 |
| Opus 5 | $0.01123 | $0.01123 |
| Sonnet 5 | $0.00449 | $0.00449 |
| Haiku 4.5 | $0.00225 | $0.00225 |
Grade A, and why
aiwaku.nvim context.instructions.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 yesterday.
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 — 203 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Context instructions (applies to all files)
Purpose
- Quickly capture short, actionable lessons when code generation and review misalign.
Scope
- Applies to all contributors and automated agents.
When to apply
- Whenever review findings, validation failures, or reviewer feedback differ from generated code, or a recurring mistake appears.
Process (short)
-
Detect
- One-line symptom: where (file/PR) and why it's a problem.
-
Review and verify
- Always perform a focused review of the generated change and run the repository's existing validation commands before finalizing changes.
- If the repository does not define dedicated automation, do a manual diff review plus at least one concrete targeted verification step for the change (for example,
git diff --checkfor docs/config edits or a load-time smoke check for Neovim Lua changes).
-
Record
- Add a short entry under .github/instructions/ including:
- What: one-line summary
- Where: file paths and PR number
- Diff/fix: paste the failing diff or the minimal fix
- Why: brief explanation of the root cause
- Verification: one-line check to confirm the fix
- Keep entries concise and code-linked.
- Add a short entry under .github/instructions/ including:
-
Update PR/commit
- Append the diff/fix and a short description to the PR body or commit message, following the Git workflow conventions in
.github/copilot-instructions.md, and reference the instruction entry.
- Append the diff/fix and a short description to the PR body or commit message, following the Git workflow conventions in
-
Automated agents
- If an automated agent generates code, it must review the result and verify it with the repository's available validation steps before finalizing changes.
- If generated code produces errors (syntax/runtime/test failures) or requires iterative guidance (e.g., re-prompting, partial fixes) to reach a correct result, record:
- Error: concise error messages or failing test names/log snippets
- Attempts: prompts or steps used to guide/fix the code
- Final diff/fix: the minimal changes that fixed the problem
- Verification: how the fix was validated (tests run, lint, manual check)
- Append the diff/fix and the above details to the instruction entry and re-run the relevant review and validation steps; do not merge until verification passes.
- Store a concise memory/fact for future automation when relevant.
Verification
- Re-run the relevant review and validation steps, confirm the generated code matches the reviewed code, and mark the instruction entry verified.
Tone
- Short, example-driven, and anchored to file paths. Do not open issues by default — record and verify first.
These instructions exist to reduce repeated review failures by ensuring diffs, fixes, and explanations are preserved and discoverable.
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.
- yesterday First seen · 203 lines · 2,247 tokens per session scan A 0c6bf3f1388e
aiwaku.nvim context.instructions.md is an instructions file published in the GitHub repository juhaku/aiwaku.nvim (11 stars, last pushed 2mo ago), licensed MIT. It adds 2,247 tokens to every session, about $0.0112 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 2026-09-02.
Other instructions, from other repositories
fff AGENTS.md
AGENTS.md instructions for dmtrKovalenko/fff, covering to clankers, development commands, building, testing and development tools and code quality.
agentic.nvim AGENTS.md
Instructions for carlos-algms/agentic.nvim, covering agents guide, nested instructions, skill driven, domain glossary — lazy read and architectural decisions (adrs) — optional read.
agentic.nvim CLAUDE.md
Instructions for carlos-algms/agentic.nvim: CRITICAL See @AGENTS.md for Instructions, it's the source of truth for all my instructions, no exceptions, it should be read and be part of your context before any other action.
review.nvim CLAUDE.md
Instructions for vuki656/review.nvim, covering claude.md, commands, lint, run all tests and run a single test file.
sidekick.nvim AGENTS.md
Instructions for folke/sidekick.nvim, covering agent cheat sheet, project overview, everyday commands, adding features and writing tests.
minuet-ai.nvim AGENTS.md
AGENTS.md instructions for milanglacier/minuet-ai.nvim, covering agents.md, coding style, writing, comments & docstrings and commit messages & communication.