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/tacoda/keystone/claude-mdgit clone --depth 1 https://github.com/tacoda/keystoneWrote 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/tacoda/keystone/claude-md)<a href="https://agentmods.dev/instructions/tacoda/keystone/claude-md"><img src="https://agentmods.dev/badge/instructions/tacoda/keystone/claude-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 | $0.01036 | $0.01036 |
| Opus 5 | $0.00518 | $0.00518 |
| Sonnet 5 | $0.00207 | $0.00207 |
| Haiku 4.5 | $0.00104 | $0.00104 |
Grade A, and why
keystone CLAUDE.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 — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Keystone — the agent charter framework
Stack
- Go 1.22+ (1.25 in mise). Single binary, single module.
- CLI: Cobra (
cmd/keystone/). Subcommands wired inroot.go. - MCP server: mark3labs/mcp-go (
internal/framework/mcp/). - YAML: gopkg.in/yaml.v3. One YAML parser, period (see
guides/idioms/go/stdlib-first). - Templates: go:embed under
internal/framework/scaffold/templates/. - Release: GoReleaser via tag push (no
gh release create— see memory).
Layout
cmd/keystone/ — CLI entrypoint + every subcommand (one file each)
internal/framework/ — agent-agnostic core
primitive/ — descriptor parse, walk, index, projection
config/ — keystone.json (ProjectConfig + HookSpec)
loader/ — policy cascade + verify
policies/ — vendored policy resolver
lockfile/ — install state
manifest/ — policy manifest schema
migrations/ — version-to-version upgrade steps
scaffold/ — keystone init / bootstrap templates
sensors/ — sensor runner registry (keystone-owned sensors only)
adapters/ — per-host projection (claudecode, cursor, …)
mcp/ — MCP server
web/ — local dashboard
.charter/ — this project's own charter (dogfood)
.claude/ — projected from .charter/ — do not hand-edit
Common commands
go build ./cmd/keystone # build the CLI
go install ./cmd/keystone # install on PATH (refresh after every Frontmatter or CLI change — hooks call the installed binary)
go test ./... # all tests
go vet ./... # static check (must pass before commit)
keystone index # regenerate .charter/INDEX*.json
keystone project # regenerate .claude/* + merge hooks into settings
keystone watch # long-running: re-project on every guide/sensor/agent save (300ms debounce)
keystone verify # policy + cascade check (auto-fires pre/post-verify hooks)
keystone signal fire <name> # dispatch primitives subscribed (on:) to a signal (hook fire = alias)
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 · 82 lines · 1,036 tokens per session scan A bedc6da9c543
keystone CLAUDE.md is an instructions file published in the GitHub repository tacoda/keystone (44 stars, last pushed 2mo ago), licensed MIT. It adds 1,036 tokens to every session, about $0.0052 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 2026-09-03.
Other instructions, from other repositories
dingtalk-workspace-cli AGENTS.md
AGENTS.md instructions for DingTalk-Real-AI/dingtalk-workspace-cli, covering repository agent guide, build and test, command framework declaration, flag / help / schema homology and agent schema contract.
azure-sdk-for-go go-code.instructions.md
Instructions for Azure/azure-sdk-for-go: All code should follow the guidelines from the Azure Go SDK Guidelines. This document is a summary of the most important guidelines to follow when contributing to the Azure Go SDK.
unraid-management-agent copilot-instructions.md
Copilot instructions for ruaan-deysel/unraid-management-agent, covering copilot instructions, copilot workflow, path-specific instructions, reusable prompts and quick commands.
coral CLAUDE.md
Instructions for cdknorow/coral, covering claude.md - coral go, mission, testing, go unit tests and legacy parity tools (historical reference).
agentcat-go-sdk CLAUDE.md
Claude Code instructions for agentcathq/agentcat-go-sdk, covering claude.md, what this is, build & test commands, module layout and architecture.
oastools GEMINI.md
Instructions for erraggy/oastools, covering gemini.md, project overview, building and running, key commands and development conventions.