MCP App StoreMCP App Store
from internta27
Multi-expert scientific agent for drug discovery and biomedical research that instantiates domain specialists, integrates multimodal evidence, and generates ana
E.Copi is a multi-agent scientific system focused on drug discovery and biomedical research. Given a research question, candidate drug combinations, or molecular structures (SMILES), it automatically instantiates specialist agents (pharmacology, pharmacokinetics, bioinformatics, toxicology, clinical, computational chemistry), orchestrates evidence flows across molecular, network, clinical and literature sources, performs mechanism reasoning and risk assessment, and outputs both structured reports and Chinese two-person podcast scripts/audio.
Use this skill when the user asks for deep, multidisciplinary analysis of drug candidates, mechanism-of-action hypotheses, combination therapy assessments, risk/safety evaluations, or when you want a narrative-friendly Chinese podcast-style summary of scientific findings. Trigger on drug discovery, mechanism analysis, risk assessment, or requests for expert-collated biomedical evidence.
Best used in environments with shell/script capability and web access for literature/DB queries. Works with agent platforms that can call scripts (e.g., Claude/Copilot/agents that support bash/script execution).
E.Copi is a multi-expert scientific agent for drug discovery and biomedical research that instantiates domain specialists (pharmacology, toxicology, bioinformatics etc.) and generates Chinese-language podcast dialogues via ListenHub API. The SKILL.md is extensive and detailed but monolithic. All three scripts failed: create-podcast.sh and get-speakers.sh depend on a missing lib.sh shared library not included in the bundle, while upload-wechat.sh requires WECHAT_APPID/WECHAT_APPSECRET env vars and an audio file path. No security concerns found — scripts use env vars for secrets and make documented API calls to ListenHub and WeChat.
lib.sh (referenced but not bundled in scripts)Well-conceived scientific multi-agent concept. Main issues: (1) critical missing dependency lib.sh makes 2/3 scripts dead on arrival, (2) SKILL.md is a wall of text without progressive disclosure, (3) Chinese-language focus limits broad applicability. No malicious patterns detected.
Seq Wrangler
Sequence data QC, trimming, alignment, and BAM processing pipeline helper — wraps FastQC, fastp/Trimmomatic, BWA/Bowtie2/Minimap2, and SAMtools for automated re
Maestro Orchestrate
Orchestrates complex engineering workflows with enforced design and approval gates, child-agent delegation, and session state management.
Build Unplanned Feature
Orchestrates ad-hoc feature development: requirement gathering, planning, implementation and validation across task-planning, implementation, and validation loo
AI Pair Collaboration
Orchestrates a small heterogeneous team of AI agents (author/developer + Codex reviewer + Gemini reviewer) to produce and review content or code in a coordinate
ToolUniverse Gateway
Unified agent gateway to 1000+ scientific tools (bioinformatics, drug discovery, genomics) via Harvard's ToolUniverse, returning JSON for downstream workflows.
Box0 (b0)
Run and coordinate parallel AI sub-agents locally or on a server to delegate reviews, tests, research, and other specialist tasks.
MatchMS (mass spectrometry processing)
Mass spectrometry data processing: import/export mzML/MGF/MSP, filter and normalize spectra, compute spectral similarity (cosine/modified cosine), and build rep
Turkey Build
Multi-agent app builder and orchestrator: bootstraps projects, creates feature branches, and runs runtime-verified builds with a PM conductor and QA gates.