Automated summary
Drives an agent that proposes lattice-site occupancy swaps at fixed composition to lower the relaxed energy of an atomistic configuration.
Strengths
- Enforces a strict JSON response protocol (index pairs to exchange plus a stop flag) that an automated optimizer can consume directly.
- States checkable constraints up front: preserved species counts, lattice sites only, frozen framework atoms, X treated as a vacancy.
- Gives a phased search strategy (diversify, exploit, small local moves) with explicit stopping criteria.
- Calls out common failure modes: swapping identical species, repeating occupancies already in the history.
Limitations
- Tightly coupled to the Agenticoptimizer ecosystem and a config.toml with specific keys (swap_params.lattice_sites, atom_swaps, ratio).
- Assumes the agent receives current occupancy, energy and history, but offers no way to estimate a swap's effect before relaxation.
- Covers fixed-composition permutation only; no guidance on changing cell size, framework chemistry or target ratio.
Use it when you must pick which lattice sites to exchange in order to reduce the energy of a fixed-composition configuration inside an agentic optimization loop.
Avoid it if the task requires altering composition, cell size or framework chemistry, or if occupancy and energy are not supplied at each step.
Security analysis
SafeThe skill only instructs an agent to propose lattice-site swaps and return JSON; it does not request shell execution, network access, file writes, secret handling, or destructive operations. No tools are declared, so execution risk is minimal.
No concerns found
How this analysis is madeExamples
Current occupancy (lattice sites 0-15): Na Na Na X Na Na X Na Na Na Na X Na Na Na Na. Energy: -412.37 eV, previous best -412.10 eV. Composition must stay 66% Na / 34% X. Propose the next site swaps as JSON.Energy has not improved in the last 5 steps and the configuration still shows a block of four adjacent Na sites at indices 4-7. Suggest swaps that break that clustering without changing species counts, and return only the JSON move.History shows no energy improvement for 8 consecutive steps and the current occupancy already appeared earlier in the run. Tell me if we should stop optimizing, using the JSON done response.name: atom-site-swap description: >- Propose Na/X (or configured) lattice-site swaps to lower the relaxed energy of a fixed-composition configuration. Use when optimizing site occupancy in Agenticoptimizer, reading energy feedback, or choosing which atoms to exchange.
Atomistic site-swap agent
Goal
Find a low-energy arrangement of species on the lattice sites at a fixed composition (ratios from config.toml). This replaces genetic-algorithm operators: you choose which sites to swap.
Do not change the number of each species. Only permute occupations on lattice sites.
Domain rules
- Lattice sites come from
swap_params.lattice_sites(e.g. all Na positions in the supercell). - Allowed species are
swap_params.atom_swaps(e.g.NaandX). Xmeans a vacancy. It is removed before relaxation; do not invent other elements.- Composition must stay at
swap_params.ratio(e.g.[0.66, 0.34]→ ~66% first species, ~34% second). - Framework atoms (everything that is not a lattice site) are fixed — never swap them.
- Lower relaxed total energy is better. Prefer swaps that break unfavorable clustering or improve mixing when energy stalls.
How to propose a move
Each step you receive:
- current occupancy as a list of symbols, one per lattice site index
0 .. N-1 - current energy (eV) and recent history
- counts of each species (must be preserved)
Respond with only a JSON object (no markdown fences, no commentary):
{"swaps": [[i, j], [k, l]], "rationale": "short reason"}
- Each
[i, j]exchanges the species on sitesiandj. - Prefer 1–4 swaps per step; more is allowed if justified.
- Only swap pairs that currently have different species (otherwise the move is a no-op).
- Site indices must be in range and refer to lattice sites only.
If you believe the current configuration is optimal enough to stop:
{"swaps": [], "done": true, "rationale": "why stop"}
Strategy hints
- Early: diversify — break large same-species clusters; sample different local arrangements.
- Mid: exploit — refine neighborhoods that previously lowered energy.
- Late: small local swaps only; stop when energy no longer improves for several steps.
- Never propose a move that changes species counts.
- Avoid repeating an occupancy that already appears in the history.
What you must not do
- Do not change cell size, framework chemistry, or the target ratio.
- Do not request DFT settings changes mid-run; relaxation params come from config.
- Do not output prose outside the JSON object.
From the file to a first result
The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.
View the source and supporting filesThis download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.
1. Download the file
Download SKILL.mdFrom the root of a test project, save the file as .skill-downloads/atomistic-site-swap-mrwakezh/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.
Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.
mkdir -p '.skill-downloads/atomistic-site-swap-mrwakezh' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/atomistic-site-swap-mrwakezh/download' --output '.skill-downloads/atomistic-site-swap-mrwakezh/SKILL.md.part' && mv '.skill-downloads/atomistic-site-swap-mrwakezh/SKILL.md.part' '.skill-downloads/atomistic-site-swap-mrwakezh/SKILL.md'2. Check prerequisites
Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.
No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.
The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.
3. Load the skill
After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.
Claude Code
test -f '.skill-downloads/atomistic-site-swap-mrwakezh/SKILL.md' && test -s '.skill-downloads/atomistic-site-swap-mrwakezh/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/atom-site-swap' && cp -R '.skill-downloads/atomistic-site-swap-mrwakezh' '.claude/skills/atom-site-swap'Cursor
test -f '.skill-downloads/atomistic-site-swap-mrwakezh/SKILL.md' && test -s '.skill-downloads/atomistic-site-swap-mrwakezh/SKILL.md' && mkdir -p '.cursor/skills' && test ! -e '.cursor/skills/atom-site-swap' && cp -R '.skill-downloads/atomistic-site-swap-mrwakezh' '.cursor/skills/atom-site-swap'Codex
test -f '.skill-downloads/atomistic-site-swap-mrwakezh/SKILL.md' && test -s '.skill-downloads/atomistic-site-swap-mrwakezh/SKILL.md' && mkdir -p '.agents/skills' && test ! -e '.agents/skills/atom-site-swap' && cp -R '.skill-downloads/atomistic-site-swap-mrwakezh' '.agents/skills/atom-site-swap'OpenSkills
OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.
Official OpenSkills installationopenskills install './.skill-downloads/atomistic-site-swap-mrwakezh'openskills read 'atomistic-site-swap-mrwakezh'Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.
For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.
openskills syncManual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.
4. Try a first task
Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:
Read .skill-downloads/atomistic-site-swap-mrwakezh/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.
Create Interactive Maps with kepler.gl
Data & AI
Create interactive map visualizations and export standalone HTML using the keplergl Python package. Use to plot geospatial data from DataFrames, GeoJSON, or CSV.
Agent-Driven Parameter Optimization
Data & AI
Guide for optimizing configurable systems with a measurable objective, using bounded trials and evidence-based recommendations.
ICP Visitor Outreach
Data & AI
Identify signed-up users matching the ideal customer profile but stalled, and draft personalized outreach to re-engage them.