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/sandovaldavid/kioku/agents-mdgit clone --depth 1 https://github.com/sandovaldavid/kiokuWrote 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/sandovaldavid/kioku/agents-md)<a href="https://agentmods.dev/instructions/sandovaldavid/kioku/agents-md"><img src="https://agentmods.dev/badge/instructions/sandovaldavid/kioku/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 5.1 | $0.02106 | $0.02106 |
| Opus 5 | $0.01053 | $0.01053 |
| Sonnet 5 | $0.00421 | $0.00421 |
| Haiku 4.5 | $0.00211 | $0.00211 |
Grade A, and why
kioku 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 5d 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 — 194 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Kioku agent operating manual
This file defines how coding agents must work in sandovaldavid/kioku. It is an operational guide for the current develop branch, not a roadmap or historical report.
Repository scope
Kioku is a .NET 10 Model Context Protocol server that reads and writes an Obsidian vault. The server supports local stdio and authenticated Streamable HTTP transports. The optional Obsidian bridge plugin is maintained and released independently in sandovaldavid/kioku-obsidian.
The server accesses the vault filesystem directly. Obsidian does not need to be open for core note, search, project, session, indexing, or coordination operations. A running Obsidian application and plugin are required only for operations registered through the optional bridge and plugin capability groups.
This repository owns:
- the MCP server, contracts, prompts, resources, and integrations;
- vault indexing, retrieval, filesystem policy, work sessions, and engineering workflows;
- server packaging, Docker deployment, CI, releases, retrieval evaluation, and public documentation.
This repository does not own the Obsidian plugin source or its release workflow.
Source-of-truth order
When statements disagree, use this precedence:
- current code and configuration on the target branch;
- executable tests and generated public contracts;
- maintained operational documentation;
- merged pull requests and closed issues as historical evidence only;
- open issues, proposals, plans, and comments as non-implemented intent.
Never describe an issue, roadmap item, proposed architecture, benchmark expectation, or PR body as implemented unless the target branch contains the corresponding code or generated contract.
Current verified snapshot
- Target framework:
net10.0, configured inDirectory.Build.props. - Solution:
Kioku.slnx. - Latest published server package:
3.1.2. The integration branch can contain unreleased changes beyond that release. - MCP C# SDK packages:
ModelContextProtocolandModelContextProtocol.AspNetCore2.2.0. - MCP protocol baseline:
2026-07-28. - Streamable HTTP runs explicitly with
Stateless = true; localstdioremains supported. - Default MCP profile: 45 tools.
- All-capabilities profile: 78 tools.
- Engineering projects support first-class
specdocuments throughcreate_engineering_spec;create_implementation_plancan link a same-project spec, andget_project_contextacceptsspec/specsaliases. - New project scaffolds create
decisions,bugs,specs,plans,knowledge,sessions, andbacklogeagerly;dailyandticketsremain supported optional/lazy workflows. - Optional groups disabled by default:
research,generation,css,assets,bridge,plugin, andcoordination. - Supported client integrations: Claude Code, Codex, OpenCode, GitHub Copilot, and Antigravity via native client MCP registration.
- The generated
docs/commands-reference.mdis authoritative for tool names, schemas, annotations, prompts, resources, and profile counts. - The generated
docs/configuration-reference.mdis authoritative for public process configuration. - Markdown files in the vault are the durable source of truth. Runtime indexes and the embeddings cache are derived data; there are no database migrations to maintain.
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.
- 5d ago First seen · 194 lines · 2,106 tokens per session scan A 840c08afdf7a
kioku AGENTS.md is an instructions file published in the GitHub repository sandovaldavid/kioku (0 stars, last pushed 7d ago), licensed MIT. It adds 2,106 tokens to every session, about $0.0105 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 2026-08-31.
Other instructions, from other repositories
obsidian-mcp-server AGENTS.md
AGENTS.md instructions for cyanheads/obsidian-mcp-server, covering agent protocol, what's next?, core rules, patterns and tool — obsidianlisttags.
obsidian-mcp-plugin CLAUDE.md
Claude Code instructions for aaronsb/obsidian-mcp-plugin, covering claude development guidelines for obsidian mcp plugin, project context, code quality guidelines, solid principles application and architecture patterns.
enquire-mcp AGENTS.md
AGENTS.md instructions for oomkapwn/enquire-mcp, covering agents.md, tl;dr, architecture (5-minute orientation), conventions and typescript.
Connapse CLAUDE.md
Claude Code instructions for Destrayon/Connapse, covering claude.md, build & run, build, run web app (requires backing services) and full stack via docker.
MCP-Plugin-dotnet CLAUDE.md
Claude Code instructions for IvanMurzak/MCP-Plugin-dotnet, covering claude.md, what this is, build / run, critical invariants and find detail in.
obsidian-mcp-server copilot-instructions.md
Copilot instructions for Vasallo94/obsidian-mcp-server, covering development guidelines, core development rules, development philosophy, coding best practices and system architecture.