← Back to Skills Marketplace
avale-slai

Loomlensai

by AVale-SLAI · GitHub ↗ · v1.0.3 · MIT-0
cross-platform ⚠ suspicious
116
Downloads
0
Stars
0
Active Installs
4
Versions
Install in OpenClaw
/install loomlensai
Description
Estimate the cost of any AI prompt across 19+ models before you run it. Works with OpenAI, Anthropic, Google, DeepSeek, xAI, MiniMax, and local models.
README (SKILL.md)

LoomLens AI™ — Prompt Cost Estimator

What It Does

Estimates the exact cost in USD of running any prompt across 19+ AI models — before you spend a single token. Shows input tokens, output tokens, cached tokens, and cost per model for low/mid/high complexity bands.

When to Use

  • Before running a large prompt across multiple agents
  • When optimizing for cost in production pipelines
  • When comparing models for a new use case
  • After prompt changes to check budget impact

Syntax

In chat, type:

/loomlens Estimate the cost of this prompt across GPT-4o and Claude Sonnet 4

Free Tier

3 estimates/day free with any Signalloom API key (no card required).

Get your free key: https://signalloomai.com/signup

Paid: $0.10/estimate (100 LoomTokens). LoomLens Pro: $15/mo for 1,000 estimates.

Output

Returns cost breakdown per model:

  • Input / Output / Cached token estimates
  • Low / Mid / High cost bands
  • Cheapest model recommendation
  • Savings vs most expensive option

Examples

/loomlens "You are an expert radiologist. Analyze this chest X-ray..." --agents 3 --runtime 5
/loomlens "Summarize this 10-page document" --models openai/gpt-4o anthropic/claude-haiku-4
/loomlens "Code review: find all security issues in this diff" --complexity high
Usage Guidance
This skill is plausible for estimating prompt costs but has several red flags you should understand before installing: - Missing binary: The skill requires a 'loomlens' CLI but the package does not include it or install it. Do not download that binary from an untrusted source — ask the publisher where to get the official loomlens binary or verify its integrity (checksums, official release page). - Telemetry: install.sh makes an unauthenticated POST to api.signalloomai.com to record installs. That sends only the skill slug/version/source, but if you prefer no telemetry, review or remove that curl line before running the installer. - Shell changes: install.sh will symlink into ~/.openclaw/skills and append an export PATH line into your ~/.zshrc if it doesn't find the literal string. Back up your shell config before running, and consider adjusting the PATH change to match your shell (bash, fish) manually. - Incomplete/ inconsistent package: package.json advertises a bin script that is not present and version numbers don't match. Prefer to obtain the skill from a source with a homepage or repository you can audit. Recommended actions: inspect the contents of the skill directory, open install.sh and package.json (you have them), and verify the official loomlens binary source (homepage/repo). If you proceed, run the installer in a disposable environment or container, or manually perform only the symlink/PATH actions you understand. If you need higher assurance, ask the skill author for a canonical release URL, checksums, and a privacy policy explaining what data is sent to signalloomai.
Capability Analysis
Type: OpenClaw Skill Name: loomlensai Version: 1.0.3 The skill bundle contains an installation script (install.sh) that performs background telemetry and modifies sensitive shell configuration files. Specifically, install.sh sends a POST request to api.signalloomai.com to track installations and appends a PATH modification to the user's ~/.zshrc without explicit consent. While these behaviors are common in some software installers, they constitute risky network and file system access that are not strictly necessary for the tool's stated purpose of prompt cost estimation. Additionally, there is a functional discrepancy where the script modifies the PATH to include ~/.local/bin but fails to actually install the referenced 'loomlens' binary into that directory.
Capability Assessment
Purpose & Capability
The SKILL.md describes a prompt-cost estimator across many models, which coheres with requiring a 'loomlens' CLI. However the package bundle does not include the loomlens binary or its installer, and package.json points to a ./scripts/loomlens entry that is not present. Version strings also mismatch between package.json (1.0.1), install.sh (1.0.2), and the registry (1.0.3). These gaps mean the skill cannot function without a separate loomlens binary from an external source; that missing dependency is an unexplained friction point.
Instruction Scope
SKILL.md instructions are narrowly scoped to cost estimation and invoking /loomlens. The included install.sh, however, modifies the user's environment (symlinks into ~/.openclaw/skills, appends an export PATH line to ~/.zshrc unconditionally if a literal string isn't found) and prints instructions to set SL_API_KEY. The installer also sends an unauthenticated analytics POST to https://api.signalloomai.com/v1/analytics/install. No other unrelated file reads or env access are requested in the SKILL.md.
Install Mechanism
There is no formal install spec in the registry metadata; installation is instruction-only but ships an install.sh that writes a symlink into ~/.openclaw/skills and appends to ~/.zshrc. The install script does not install the required 'loomlens' binary; it only configures the skill directory and PATH. The package.json advertises a bin path that is missing. These issues create a confusing, incomplete install process and raise risk if the user must fetch the missing binary from an unknown source.
Credentials
The skill declares no required environment variables, yet install.sh and the SKILL.md mention an SL_API_KEY (Signalloom) and provide a signup URL. The installer also sends an analytics POST (no auth) to signalloomai. Requesting a single optional API key for a telemetry/billing service is plausible, but the lack of declared env requirements and the presence of telemetry are inconsistent and should be made explicit to users.
Persistence & Privilege
The skill does not request 'always: true' and allows normal autonomous invocation. The installer creates a persistent symlink in ~/.openclaw/skills and appends PATH exports to ~/.zshrc, which are reasonable for a CLI-based skill but are modifications to user shell config that should be highlighted to users. There is no evidence it attempts to change other skills' configurations.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install loomlensai
  3. After installation, invoke the skill by name or use /loomlensai
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.3
- Updated install.sh script (details not specified). - No changes to functionality or documentation content.
v1.0.2
- Initial package manifest (package.json) added for LoomLens AI skill. - Lays groundwork for dependency and environment management.
v1.0.1
- Updated SKILL.md to include Openclaw metadata, such as emoji and requirements. - Revised installation and invocation instructions for simpler in-chat usage. - Removed explicit Signalloom API mention from the description and install section. - Improved clarity in syntax and usage examples. - No code or functionality changes; documentation only.
v1.0.0
LoomLens AI 1.0.0 – Initial Release - Instantly estimates USD cost for any prompt across 19+ AI models before running. - Supports OpenAI, Anthropic, Google, DeepSeek, xAI, MiniMax, and local models. - Displays token breakdown (input/output/cached) and cost per complexity level. - Identifies cheapest model and savings vs the most expensive option. - Requires a free Signalloom API key; free tier includes 3 daily estimates.
Metadata
Slug loomlensai
Version 1.0.3
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 4
Frequently Asked Questions

What is Loomlensai?

Estimate the cost of any AI prompt across 19+ models before you run it. Works with OpenAI, Anthropic, Google, DeepSeek, xAI, MiniMax, and local models. It is an AI Agent Skill for Claude Code / OpenClaw, with 116 downloads so far.

How do I install Loomlensai?

Run "/install loomlensai" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Loomlensai free?

Yes, Loomlensai is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Loomlensai support?

Loomlensai is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Loomlensai?

It is built and maintained by AVale-SLAI (@avale-slai); the current version is v1.0.3.

💬 Comments