What is lore-oss CLAUDE.md?

Repository instructions and shared memory for Claude Code, an AI coding assistant. It describes the project’s architecture, coding conventions, team decisions, task history, and context-search services.

It helps Claude Code find project context, search team memory, follow conventions, and work with delegated tasks, GitHub Issues, and the project’s context server.

It gives the assistant the background needed to make changes consistent with the organization and project. This reduces repeated explanations and decisions based only on the current files.

How re-cinq/lore-oss configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to re-cinq/lore-oss. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/re-cinq/lore-oss/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/re-cinq/lore-oss

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/re-cinq/lore-oss/claude-md/github.svg)](https://agentmods.dev/instructions/re-cinq/lore-oss/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/re-cinq/lore-oss/claude-md"><img src="https://agentmods.dev/badge/instructions/re-cinq/lore-oss/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/re-cinq/lore-oss/claude-md"><img src="https://agentmods.dev/badge/instructions/re-cinq/lore-oss/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 3,709 This file is loaded in full into every session.

When invoked 3,709 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.03709	$0.03709
Opus 5.5	$0.01484	$0.01484
Sonnet 5.5	$0.00742	$0.00742
Haiku 4.5	$0.00371	$0.00371

Measured 2mo ago against content hash 77ee8f119f4f, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

lore-oss 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 2mo 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.

CLAUDE.md · 321 lines

How it starts

The opening of the file, as written. The whole thing — 321 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Lore

Shared context infrastructure for Claude Code. One install command gives developers full org awareness — conventions, ADRs, team patterns, PR history, and task state.

Architecture

MCP server (mcp-server/src/index.ts + routes.ts): TypeScript, serves context to Claude Code via MCP protocol. Dual transport: stdio for local (Phase 0), Streamable HTTP for GKE (Phase 1). Three core tools: assemble_context, search_context, search_memory. Pipeline delegation, local task runner, and 30+ tools total.

Vector store: PostgreSQL + pgvector via CloudNativePG on GKE. Schema-per-team isolation. HNSW indexes for vector search, GIN for BM25 keyword search. Hybrid search via Reciprocal Rank Fusion. Embeddings from Vertex AI text-embedding-005 (768 dimensions).

Cluster agents: Lore Agent service on GKE processes pipeline tasks via direct Anthropic API calls (simple tasks) or headless Claude Code (complex tasks). Developers delegate through the Lore MCP server, never directly.

Observability: OpenTelemetry traces + metrics → Cloud Monitoring. Gap signal goes to Graphiti episodes in Phase 3.

Task tracking: Pipeline tasks via Lore MCP + GitHub Issues.

Code Conventions

TypeScript for the MCP server. ESM modules, strict mode, ES2022 target. Zod for input validation on all MCP tools. Return errors as text in MCP responses, never throw.

Python for glue scripts (lore-gen-constitution). Keep them short (<100 lines). Handle missing tools gracefully with clear error messages.

Bash for install.sh, lore-doctor, infra scripts. Must be idempotent — safe to re-run. Prefix output with [lore]. Exit 0 on success, 1 on failure.

Helm charts for K8s deployments (Lore Agent, MCP server). Values files should have sane defaults. No hardcoded secrets — use K8s Secrets.

No long-lived credentials anywhere. Workload Identity on GKE, gcloud auth for local dev.

Key Components

mcp-server/ — the MCP server (TypeScript)
mcp-server/src/routes.ts — HTTP API route handlers (extracted from index.ts)
mcp-server/src/github-client.ts — consolidated GitHub auth (App + token fallback)
mcp-server/src/local-runner.ts — local task runner (worktrees, background Claude Code)
scripts/ — install.sh, lore-doctor, lore-init, glue scripts
scripts/infra/ — setup-db.sh, setup-schedulers.sh, generate-embeddings.sh
scripts/klaus-prompts/ — standing instructions for agents (legacy, migrating to lore-agent)
.claude/skills/ — platform skills (lore-feature, lore-pr, lore-init)
terraform/modules/ — K8s manifests, Helm charts (lore-db, gke-mcp)
docker/claude-runner/ — ephemeral container for Claude Code execution in K8s Jobs
terraform/modules/gke-mcp/loretask-crd/ — LoreTask CRD, RBAC, controller deployment
specs/ — speckit artifacts (spec, plan, tasks, research, contracts)
adrs/ — architecture decision records (MADR format)
teams/ — per-team CLAUDE.md files
agent/src/platform.ts — CodePlatform interface (branch, commit, PR, issue, repo content, PR details)
agent/src/github.ts — GitHubPlatform implementation (only file importing Octokit)
web-ui/src/lib/github.ts — GitHub App client for web-ui (PR status fetching)
web-ui/src/app/pipeline/[id]/TaskLogs.tsx — live Job log viewer (polls every 5s)
web-ui/src/app/pipeline/[id]/PRStatusCard.tsx — live PR status card
agent/src/jobs/loretask-watcher.ts — polls LoreTasks, creates PRs, triggers auto-review
mcp-server/src/context-assembly.ts — context assembly with YAML templates
mcp-server/templates/ — YAML context assembly templates (default, review, implementation, research)
mcp-server/src/repo-validation.ts — deterministic validation (lint/typecheck detection for Node/Go/Python/Rust)
mcp-server/src/repo-validation-cli.ts — CLI wrapper for validation in K8s Job pods
scripts/slack-app-manifest.yaml — Slack app manifest for /lore slash command
agent/src/lib/episode-writer.ts — shared episode writer with Haiku-driven auto-curation
agent/src/jobs/memory-lifecycle.ts — importance decay (eviction) + fact consolidation (pattern extraction)
mcp-server/src/session-tracker.ts — passive session tracking (tool calls, ring buffer, exit dump)
evals/ — PromptFoo eval configs per team

Read the full file on GitHub · 321 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.

2mo ago First seen · 321 lines · 3,709 tokens per session scan A 77ee8f119f4f

Subscribe to this mod's changes

lore-oss CLAUDE.md is an instructions file published in the GitHub repository re-cinq/lore-oss (12 stars, last pushed 7mo ago), licensed Apache-2.0. It adds 3,709 tokens to every session, about $0.0148 per session on Opus 5.5. 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.

Memory and context

lore-oss CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
lore-oss CLAUDE.mdre-cinq/lore-oss	Instructions file	12	7mo ago	Apache-2.0	A	3,709	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	yesterday	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

lore-oss: Instructions file for Claude Code