agentmods.dev

aperture_lab_optimizer

aperture_lab_optimizer is a skill for Claude Code, Codex from tomismeta/aperture. It costs 59 tokens per session (613 once invoked), scanned A, original, MIT.

A narrowly scoped optimizer for Aperture Lab’s fixed calibration corpus, meaning a stable set of examples used to measure changes. It edits only approved semantic or importer files and returns a structured result after running required checks.

In plain words
What is it for?
Use it to reduce corrected mismatches in the calibration set, keep invariant mismatches at zero, apply small structural edits, and report whether the required gates passed.
Why use it?
It limits optimization to measured, reviewable changes and helps prevent edits to planning, continuity, product-interface, or live AI behavior.

Skill for Claude CodeCodex

Which agent this was written for is unclear — built for openclaw. Also seen: built for openclaw.

not rated 24repo today A scan SkillSpector: pass 59 tokens original MIT

Good fit Use it to reduce corrected mismatches in the calibration set, keep invariant mismatches at zero, apply small structural edits, and report whether the required gates passed.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/tomismeta/aperture/aperture-lab-optimizer
View source ↗ tomismeta/aperture
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.

Any agent
npx skills add tomismeta/aperture --skill aperture-lab-optimizer
Clone the repo
git clone --depth 1 https://github.com/tomismeta/aperture

Made for: Claude Code, Codex.

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.

agentmods badge for aperture_lab_optimizer

README.md
[![agentmods](https://agentmods.dev/badge/skills/tomismeta/aperture/aperture-lab-optimizer.svg)](https://agentmods.dev/skills/tomismeta/aperture/aperture-lab-optimizer)
Your own site
<a href="https://agentmods.dev/skills/tomismeta/aperture/aperture-lab-optimizer"><img src="https://agentmods.dev/badge/skills/tomismeta/aperture/aperture-lab-optimizer.svg" alt="Measured on agentmods" height="20"></a>
Per session 59 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 613 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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 5.1 $0.00059 $0.00613
Opus 5 $0.00030 $0.00307
Sonnet 5 $0.00012 $0.00123
Haiku 4.5 $0.00006 $0.00061

Measured 8d ago against content hash a87e00a5eecc, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

aperture_lab_optimizer 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 8d 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.

skills/aperture-lab-optimizer/SKILL.md · 88 lines

How it starts

The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Aperture Lab F-Stop Optimizer

Use this skill when you are the optimizer model inside Aperture Lab F-Stop.

Your role is narrow:

  • read the optimization prompt from stdin or the provided task input
  • read the repo-local program/config it points to
  • edit only the allowed semantic/importer files
  • run the required gates
  • return a structured JSON outcome

Do not:

  • edit planner or continuity files
  • add AI to the live runtime path
  • change product-shell or TUI behavior
  • create commits or switch branches
  • write release notes or bump packages

Load First

Read these before making changes:

  • packages/lab/research/autoresearch-program.md
  • packages/lab/research/autoresearch-config.json

These are the source of truth for:

  • allowed edit paths
  • forbidden edit paths
  • evaluation commands
  • optimization goals

Main Rule

Optimize against the frozen calibration corpus, not against raw reviewer noise.

That means:

  • reduce corrected mismatches
  • keep invariant mismatches at zero
  • prefer narrow edits over broad rewrites
  • prefer structural rules over one-off titles or phrase literals
  • treat the harness evaluation outputs as the source of truth

Expected Workflow

  1. Read the optimization prompt and identify the top mismatch clusters.
  2. Edit only the allowed semantic/importer files.
  3. Run the required evaluation commands from the prompt.
  4. Prefer structural generalizations over single-title or exact-phrase special cases unless multiple promoted examples clearly justify the phrase.
  5. Use the mismatch counts from the commands you actually ran; do not invent or round them.
  6. If the calibration improves but another gate fails, say that explicitly instead of calling the patch a semantic regression.
  7. Stop if the calibration score does not improve or if invariants regress.
  8. Return exactly one JSON object with:
    • action: patched or no_patch
    • summary
    • reasons
    • recommendedFiles
    • changedFiles
    • commandsRun
    • beforeMismatchCount
    • afterMismatchCount
    • judgmentBattle
    • releaseCheck

Read the full file on GitHub · 88 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. 8d ago First seen · 88 lines · 59 tokens per session scan A a87e00a5eecc

Subscribe to this mod's changes

aperture_lab_optimizer is a skill published in the GitHub repository tomismeta/aperture (24 stars, last pushed today), licensed MIT. It adds 59 tokens to every session and 613 once invoked, about $0.0003 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 2026-08-30.

Data and AITypeScriptai-agentsapproval-workflowsattention-managementcontrol-planehuman-in-the-loop
Related

Other skills, from other repositories

cuml-machine-learning

Use for GPU-accelerated machine learning on tabular data using NVIDIA cuML. Triggers when tasks involve classification, regression, clustering, dimensionality reduction, or model training on datasets.

langchain-ai/deepagents · 43 tokens

developing-genkit-js

Develop AI-powered applications using Genkit in Node.js/TypeScript. Use when the user asks about Genkit, AI agents, flows, or tools in JavaScript/TypeScript, or when encountering Genkit errors, validation issues, type errors, or API problems.

google/skills · 60 tokens

system-prompts

Write system prompts, tool docs, and agent definitions. Project tag conventions + RFC 2119 keywords + dense compression. Use when authoring or editing any prompt the model reads.

can1357/oh-my-pi · 40 tokens

tool-prompt-optimization

Optimize the description prompts an AI agent reads to learn its built-in tools (the .md files under prompts/tools/). Two halves: (1) measure how much of a prompt is already inferable from the tool's JSON parameter schema + name, to prune redundancy with evidence; (2) house authoring rules for what belongs in a tool…

can1357/oh-my-pi · 0 tokens

transformers-js

Use Transformers.js to run state-of-the-art machine learning models directly in JavaScript/TypeScript. Supports NLP (text classification, translation, summarization), computer vision (image classification, object detection), audio (speech recognition, audio classification), and multimodal tasks. Works in browsers and…

huggingface/skills · 89 tokens

ax-ai

This skill helps an LLM generate correct AI provider setup and configuration code using @ax-llm/ax. Use when the user asks about ai(), providers, models, routing, adaptive balancing, presets, embeddings, batch audio with ai.transcribe() or ai.speak(), extended thinking, context caching, or mentions…

ax-llm/ax · 100 tokens