agentmods.dev

knowns CLAUDE.md

A lightweight compatibility guide for Claude Code and other AI assistants working with Knowns, a repository memory and workflow system.

In plain words
What is it for?
Use it to initialise project context, search and retrieve Knowns content, follow repository rules, and use structured code-editing operations.
Why use it?
It directs agents to the project's canonical records and helps them find relevant information before changing code or documents.

Instructions file

241 6d ago C scan 1,788 tokens original MIT
View the source file knowns-dev/knowns
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/knowns-dev/knowns/claude-md
Clone the repo
git clone --depth 1 https://github.com/knowns-dev/knowns
Per session 1,788 This file is loaded in full into every session.
When invoked 1,788 The same file — it is already loaded in full.
Security scan C 2 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 $0.01788 $0.01788
Opus 5 $0.00894 $0.00894
Sonnet 5 $0.00358 $0.00358
Haiku 4.5 $0.00179 $0.00179

Measured 3d ago against content hash 6febedc1feac, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

knowns CLAUDE.md scanned grade C with 2 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.

Hidden instructionshighPrompt injection

Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.

<!-- rtk-instructions v2 -->

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

rtk curl <url> # Compact HTTP responses (70%)
CLAUDE.md · 178 lines

How it starts

The opening of the file, as written. The whole thing — 178 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE

Compatibility entrypoint for runtimes that auto-detect CLAUDE.md.

CRITICAL: Start with Knowns MCP initial when available. Use help("tool.*") or help("workflow.*") for domain details on demand.

Runtime Guidance

  • Knowns is the repository memory layer for humans and the AI-friendly working layer for agents.
  • MCP initial is the primary AI bootstrap: project state, tool domains, code rules, and workflow routing.
  • MCP help is the primary on-demand source for action schemas and recipes.
  • Treat this file only as a lightweight compatibility entrypoint.

Minimum Rules

  • Use Knowns as the canonical system for tasks, docs, templates, and workflow state.
  • Never manually edit Knowns-managed task or doc markdown.
  • Search first, then read only relevant docs and code.
  • Use search for discovery; use MCP retrieve tool when a workflow needs structured context with citations. Fall back to CLI knowns retrieve if MCP is unavailable.
  • For code operations, use code tool: find/symbols for structure, references/definition for navigation, rename/replace/replace_body/insert/delete for editing. Use help("code.*") or help("workflow.code-edit") for details.
  • Plan before implementation unless the user explicitly overrides that workflow.
  • Validate before considering work complete.
  • Use memory tools: memory({ action: "list" }) at session start, memory({ action: "add" }) after tasks for reusable knowledge.
  • Proactively capture durable memory when scope and durability are clear.

Quick Reference

knowns doc list --plain               # List docs
knowns task list --plain              # List tasks
knowns task <id> --plain              # View task
knowns doc "<path>" --plain --smart  # View doc
knowns search "query" --plain        # Search docs/tasks
knowns retrieve "query" --json      # Retrieve structured context pack (CLI fallback)

Read the full file on GitHub · 178 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 · 178 lines · 1,788 tokens per session scan C 6febedc1feac

Subscribe to this mod's changes

knowns CLAUDE.md is an instructions file published in the GitHub repository knowns-dev/knowns (241 stars, last pushed 6d ago), licensed MIT. It adds 1,788 tokens to every session, about $0.0089 per session on Opus 5. A static security scan graded it C with 2 findings (hidden instructions, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Memory and contextaiai-agentsai-nativeai-workflowclicontext-management
Related

Other instructions, from other repositories

claude-context-os CLAUDE.md

Instructions for Arkya-AI/claude-context-os, covering claude.md, session start, identity, rules and where things live.

Arkya-AI/claude-context-os · 687 tokens

claude-code-bridge CLAUDE.md

Claude Code instructions for harshitleads/claude-code-bridge, covering claude.md, vision and mission, current stack, architecture and code rules.

harshitleads/claude-code-bridge · 529 tokens

open-ontologies CLAUDE.md

Instructions for fabio-rovai/open-ontologies, covering open ontologies, ontology engineering workflow, generate, validate and load and reason.

fabio-rovai/open-ontologies · 5,098 tokens

vibe-coding-prompt-template backend.instructions.md

Instructions for KhazP/vibe-coding-prompt-template: Read AGENTS.md, agentdocs/techstack.md, and agentdocs/codepatterns.md.

KhazP/vibe-coding-prompt-template · 139 tokens

kin GEMINI.md

Instructions for firelock-ai/kin, covering kin: read the repository from the graph, pick the tool by what you know, trust the envelope, not the emptiness, if the tools have no graph and a working order.

firelock-ai/kin · 695 tokens

kin AGENTS.md

Instructions for firelock-ai/kin, covering agents.md and this repo's role.

firelock-ai/kin · 261 tokens