dsh-chem
Run the following command in DeepSeek Harness:
dsh plugin install TYEclipse/dsh-chem
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install TYEclipse/dsh-chem in your terminal to install the plugin; the source repository is https://github.com/TYEclipse/dsh-chem .
About this plugin
Large language models fail chemistry arithmetic in predictable ways: miscounted subscripts, dropped hydrates, confusing Co with CO, missing a dilution factor, or inverting a pH log. These are not edge cases; they surface every time a multi-step chemistry computation is involved. dsh-chem removes that class of error entirely by replacing model intuition with pure arithmetic over IUPAC standard atomic weights. Zero runtime dependencies, no network calls, no state.
Four tool families are covered: formula parsing and molar-mass calculation (nested parentheses, hydrate dots, ionic charges, case auto-correction); six-way mass / mole / particle conversions; solution concentration (molarity, preparation, C1V1=C2V2 dilution, percent concentration); and pH / pOH plus Henderson-Hasselbalch buffer math, including negative pH for concentrated acids. Every numeric anchor is generated by an independent Python oracle script and cross-validated against published textbook values.
Built for developers wiring AI agents that must return reproducible chemistry results, for chemistry-education tooling, and for any integrator who is tired of explaining why CuSO4·5H2O is 249.68 g/mol rather than 159.61.
Use Cases
- Guaranteeing correct molar-mass, dilution, and pH outputs from an AI chemistry agent
- Validating student answers in a chemistry tutoring or e-learning platform
- Replacing unreliable LLM mental math in automated lab-report or formulation pipelines
Best For
- Developers building chemistry-domain AI agents
- Product and engineering teams behind chemistry-education or e-learning tools
- Lab or industrial integrators embedding chemistry calculations into automation workflows
Related Plugins
A unified suite combining hot runtime injection, task-aware thinking-mode routing, and a graded session protocol with red-team gates to sustain model diligence across long-horizon inference.
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.