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/quentincody/interlinked-cli/claude-mdgit clone --depth 1 https://github.com/QuentinCody/interlinked-cliWhat 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 | $0.17883 | $0.17883 |
| Opus 5 | $0.08941 | $0.08941 |
| Sonnet 5 | $0.03577 | $0.03577 |
| Haiku 4.5 | $0.01788 | $0.01788 |
Grade C, and why
interlinked-cli 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 2d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
interlinked harness test "rm -rf /" # Test command against rules Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- `src/harness/__tests__/evaluator.test.ts` — destructive command blocking, sleep detection, protected files, curl-to-MCP, auto-reservations, safe command allowlist How it starts
The opening of the file, as written. The whole thing — 1,005 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
What this is for (read before changing any check, gate, or threshold)
Interlinked is a portable, per-tool-call quality and security standard for agent-written code. A local daemon sits in front of an agent's tool calls and judges each one — ideally before it reaches disk. Four goals, in priority order:
- Judge agent-written code at the moment it is written, in ANY codebase. This repo is one instance, and an unrepresentative one: single language, agent-hardened, no human legacy. Portability is the product; this tree is the test fixture.
- Ratchet the quality and security of whatever codebase it runs in. Default scope is the DIFF — the edited region of this tool call, not the file. Opt-in wider scopes: whole-file ("fix the file you touched") and whole-codebase ("bring this repo up to standard": tests, types, coverage, low complexity).
- Catch problems at the earliest point the evidence exists. PreToolUse when the proposed content is enough; PostToolUse when the file on disk or a compiler is required; trajectory when the pattern spans several calls; Stop when the only observable is "they finished without doing X". A check should declare the earliest phase its evidence supports, and sit there.
- Double as a step-level training signal for other coding agents, especially small/local ones. An agent that clears the strictest per-tool-call gates is demonstrably good — closer to rewarding every step than rewarding the final diff.
The consequence that reverses the obvious reading
A check that never fires in this repo is not dead weight, and must not be retired or demoted for being quiet. It is the part of the standard this particular (strong) agent already clears. Point it at a 7B local model or at human-written legacy code and it earns its keep.
Fire rate measures the AGENT, not the check.
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.
- 2d ago First seen · 1,005 lines · 17,883 tokens per session scan C 263480aeb02b
interlinked-cli CLAUDE.md is an instructions file published in the GitHub repository QuentinCody/interlinked-cli (13 stars, last pushed 15d ago), licensed MIT. It adds 17,883 tokens to every session, about $0.0894 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). 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.