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/alaarab/phren/claude-mdgit clone --depth 1 https://github.com/alaarab/phrenWrote 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/alaarab/phren/claude-md)<a href="https://agentmods.dev/instructions/alaarab/phren/claude-md"><img src="https://agentmods.dev/badge/instructions/alaarab/phren/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 | $0.01446 | $0.01446 |
| Opus 5 | $0.00723 | $0.00723 |
| Sonnet 5 | $0.00289 | $0.00289 |
| Haiku 4.5 | $0.00145 | $0.00145 |
Grade A, and why
phren 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 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.
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 — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Open-source knowledge layer for AI agents. Published as @phren/cli on npm.
Project Location
Source lives at ~phren. Published to npm. Starter templates are bundled in the package.
Monorepo Structure
pnpm workspace with turborepo. Shipped packages:
| Package | Path | npm | Description |
|---|---|---|---|
@phren/cli |
packages/cli/ |
Published | CLI, MCP server, data layer |
@phren/vscode |
packages/vscode/ |
VS Code Marketplace | VS Code extension |
Experimental (not published, not wired into the phren CLI):
| Package | Path | Status | Description |
|---|---|---|---|
@phren/agent |
experimental/agent/ |
Private, experimental | Standalone coding agent. Built by root pnpm build; tested by the CI agent-test job (pnpm exec turbo run test --filter=@phren/agent). Run via the built phren-agent binary — not via phren. |
Outside the pnpm workspace:
| App | Path | Description |
|---|---|---|
| phren for iOS | apps/ios/ |
Native SwiftUI app: GitHub sign-in, live-syncing view/manage of findings, notes, tasks, and the review queue straight against the store repo (serverless — GitHub REST API only). PhrenKit/ is the testable Swift package; its markdown parsers are transcriptions of the CLI's TS (fixtures generated by apps/ios/scripts/generate-fixtures.mjs from real CLI code). Xcode project generated via XcodeGen from project.yml; CI in .github/workflows/ios.yml. See apps/ios/README.md. |
Key Files
| File | Purpose |
|---|---|
packages/cli/src/index.ts |
Entry point: top-level invocation routing + MCP server with 59 tools |
packages/cli/src/entrypoint.ts |
Top-level dispatcher: resolves --help/--version/--health/MCP/manage invocations and routes via the command registry |
packages/cli/src/cli-registry.ts |
Command catalog. Single source of truth for help generation and dispatch (Command type, REGISTRY, lookupCommand, DOC_TOPICS). |
packages/cli/src/cli-help.ts |
Help formatters over the registry: formatCheatSheet, formatTopic, formatCommand, formatFullHelp |
packages/cli/src/cli-handlers.ts |
Native handlers for add, init, uninstall, status, verify, mcp-mode, hooks-mode, plus argv parsing helpers |
packages/cli/src/shared.ts |
Shared infrastructure: findPhrenPath, getProjectDirs, runtimeFile, sessionMarker |
packages/cli/src/content/ |
Content operations: finding CRUD, trust filtering, dedup, consolidation |
packages/cli/src/governance/ |
Governance: policy/access/workflow config, review queue, audit log, locks |
packages/cli/src/shared/index.ts |
FTS5 indexer: buildIndex, queryRows, @import resolution, file classification |
packages/cli/src/utils.ts |
Utilities: FTS5 sanitization, synonym expansion, keyword extraction |
packages/cli/src/init/init.ts |
phren init: configures MCP + hooks for all detected agents |
packages/cli/src/link/ |
Reconciles an existing install's machine/profile wiring, hooks, and local context |
packages/cli/src/data/access.ts |
Data layer: task CRUD, machine/profile listing, finding management |
packages/cli/src/telemetry.ts |
Opt-in usage telemetry: tool call and CLI command tracking |
packages/cli/browser/graph/ |
Shared 3D memory-viewer renderer (bundled for the web UI and the VS Code webview): scene, project navigator, contents/review pane, labels |
packages/cli/src/status.ts |
phren status: health, project, stats overview |
packages/cli/starter/global/skills/ |
The phren-* slash commands phren init provisions into ~/.phren/global/skills (and symlinks into ~/.claude/skills under the managed preset). This is the source init actually reads — setup.ts copies from here. |
packages/cli/skills/ |
Older un-prefixed copies of the same five skills plus the maintainer-only docs skill. Nothing reads this at runtime: every skill lookup resolves against the store (<phrenPath>/global/skills, <project>/skills), never the package. Edit starter/global/skills/ instead. |
packages/cli/starter/ |
Bundled starter templates (4 project types, copied to ~/.phren on init) |
packages/cli/starter/templates/ |
Project templates: python-project, monorepo, library, frontend |
docs/index.html |
GitHub Pages site |
global/CLAUDE.md |
User-level instructions shipped with the package |
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.
- 3d ago First seen · 87 lines · 1,446 tokens per session scan A 0a7afee9844c
phren CLAUDE.md is an instructions file published in the GitHub repository alaarab/phren (4 stars, last pushed 9d ago), licensed MIT. It adds 1,446 tokens to every session, about $0.0072 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
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.