lightcone-cli CLAUDE.md

Repository instructions for Lightcone CLI, a command-line execution layer for ASTRA, a research-analysis specification system. ASTRA defines what a research specification means, while Lightcone CLI runs it in a controlled environment.

In plain words
What is it for?
It helps agents work on project scaffolding, execution, environment identity and controls that keep runs isolated and reproducible.
Why use it?
It clarifies the boundary between interpreting research specifications and executing them, especially while the repository is being rebuilt.

Instructions file

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/lightconeresearch/lightcone-cli/claude-md
Clone the repo
git clone --depth 1 https://github.com/LightconeResearch/lightcone-cli
Per session 38,680 This file is loaded in full into every session.
When invoked 38,680 The same file — it is already loaded in full.
Security scan E 3 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 $0.38680 $0.38680
Opus 5 $0.19340 $0.19340
Sonnet 5 $0.07736 $0.07736
Haiku 4.5 $0.03868 $0.03868

Measured 2d ago against content hash 8a1c1504fd58, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade E, and why

lightcone-cli CLAUDE.md scanned grade E with 3 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.

Reaches for credential fileshighPrivilege escalation

SSH keys, cloud credentials, git-credentials, .npmrc, /etc/shadow: reading these is how a config file becomes a credential leak.

outright, and `~/.config/uv/uv.toml` and `~/.netrc` go missing.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

the node a ported recipe likes to `rm -rf` as a prelude (measured on

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

`--nthreads=<cpus>` (tasks block in `subprocess.wait()` with the GIL
CLAUDE.md · 2,514 lines

How it starts

The opening of the file, as written. The whole thing — 2,514 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

Project Overview

lightcone-cli is Lightcone Research's execution layer for ASTRA (Agentic Schema for Transparent Research Analysis). It ships the lc executable — an agent-agnostic CLI; it bundles no agent-specific skills, hooks, or plugins.

  • ASTRA = pure specification: schema, validation, prior insights & findings, evidence verification, helpers, minimal CLI
  • lightcone-cli = execution layer: project scaffolding, execution, environment identity, hermeticity enforcement

lightcone-cli depends on ASTRA. The astra CLI handles spec operations; the lc CLI handles execution.

That split is enforced in code, not just described: everything about what a spec means — scoping, from: references, conditional outputs, universe resolution, the recipe placeholder grammar — is answered by astra.resolve and validated by astra.validation before lc acts on it. lc's whole ASTRA surface is ten functions; see Key Invariants (layer 4).

⚠️ This repository is a clean rebuild in progress

The codebase is being re-added layer by layer on top of the design spec:

../lightcone-cli/docs/design/execution-environment.md"the locked environment is the execution environment", v6.1. Read it before adding anything — but read it as a reference, not gospel: the rebuild deliberately drifts from it as implementation teaches better answers, and where this file's Recorded decisions disagree with the spec, the decisions win. It lives in the sibling checkout for now (branch redesign_prototype), alongside its decision records (rationale, substrate tradeoffs, hermeticity enforcement, the v6 review). It stays in the sibling checkout and is dropped when the rebuild completes (decision, 2026-08): the design records are not imported into this repo's docs — the rewritten docs/ carries the current design, and this file carries the decisions.

The pre-rebuild codebase (Snakemake shim, authored Containerfiles, container: in astra.yaml, vendored dask executor plugin, WRROC export) was stripped deliberately. Functionality comes back one layer at a time, each layer landing with its own tests and dependencies — never speculatively.

Read the full file on GitHub · 2,514 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. 2d ago First seen · 2,514 lines · 38,680 tokens per session scan E 8a1c1504fd58

Subscribe to this mod's changes

lightcone-cli CLAUDE.md is an instructions file published in the GitHub repository LightconeResearch/lightcone-cli (22 stars, last pushed 7d ago), licensed BSD-3-Clause. It adds 38,680 tokens to every session, about $0.1934 per session on Opus 5. A static security scan graded it E with 3 findings (reaches for credential files, recursive force delete, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other instructions, from other repositories

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.

github/spec-kit · 7,104 tokens

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).

microsoft/vscode · 6,785 tokens

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.

openai/codex · 5,182 tokens

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.

langchain-ai/langchain · 4,345 tokens

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).

microsoft/vscode · 5,001 tokens

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.

vercel/next.js · 7,296 tokens