by research

Other Community v1.0

Skill notes

When to Use This Skill

Use this skill when you have: ✅ A protein target name or identifier (UniProt accession, gene name, PDB ID, or common name) ✅ A new design campaign starting — research is always step 1 before planning ✅ A need to characterize prior art — existing antibodies, nanobodies, or designed binders ✅ A target with unknown or partially characterized epitopes — needs hotspot identification from literature + structure

Inputs

Required: Target identifier (one of): Common name (e.g., "TNF-alpha", "PD-L1", "HER2") UniProt accession (e.g., "P01375") — most specific, preferred Gene name (e.g., "TNFSF2")

Outputs

All outputs go to the campaign directory at campaigns/{target}/campaign_{date}_{id}/research/: File · Type · Phase · Purpose scope.json · JSON · 1 · Research boundaries (target, organism, modality, depth) research_plan.json · JSON · 2 · Search strategy, priority order, time budget sources.json · JSON · 3, 7 · All retrieved sources with credibility scores

Clarification Questions

⚠️ CRITICAL: ASK THIS FIRST — confirm you have a usable target identifier and campaign directory before doing any retrieval. Target identifier (ASK THIS FIRST) — What is the target? Provide a UniProt accession if available, otherwise common name + organism. Without this, no retrieval is possible. Campaign directory — Where should research outputs land? Default is campaigns/{target}/campaign_{YYYYMMDD}_{nnn}/research/. Confirm the path or let the campaign-manager skill allocate one.

Standard Workflow

🚨 MANDATORY: USE THE 8-PHASE PIPELINE EXACTLY AS DEFINED — DO NOT COMBINE OR SKIP PHASES 🚨 The 8-phase pipeline (SCOPE → PLAN → RETRIEVE → TRIANGULATE → SYNTHESIZE → CRITIQUE → REFINE → PACKAGE) is defined in the next section. Each phase has a discrete output written to the campaign directory. After every phase, update research_progress.json before moving on. Discipline: ✅ Write the phase's JSON output before claiming the phase is complete

Fluxly profile Runtime and permissions are inferred by Fluxly static checks, not author declarations or runtime test results.
Domain
Other
Author-declared agents
No explicit declaration found; this is not inferred or tested compatibility
Static check
92 / 100 · heuristic scan, not runtime safety proof
Runtime evidence
Not run; no isolated runtime test evidence is currently recorded
Author / version / license
@001TMF · v1.0 · no license declared
Token usage
Heavy
Setup complexity
Guided setup
External API key
Not required
Operating systems
Unspecified (assume cross-platform)
Runtime requirements
Python
Detected file/system behavior
  • Read-only
  • Write / modify
  • Shell exec
Network behavior
External requests
Install commands
None (reference only)

Heads up: Tool permissions are not declared; actual access depends on the runtime.

01
Phase 1: SCOPE

Define the research boundaries before searching anything. Write `research/scope.json` to the campaign directory: Scope comes from user input and quick heuristics. Do not search databases yet.

02
Phase 2: PLAN

Create a research strategy based on the scope. Write `research/research_plan.json`: Priority: structure data first (most reliable for interface analysis), then literature, then existing binders, then homologs with known binders.

03
Phase 3: RETRIEVE

Execute all searches. Run tools in parallel where possible: `mcp__by-research__research_get_target_info` — UniProt protein info + PDB structures `mcp__by-research__research_search_prior_art` — PubMed + bioRxiv literature `mcp__by-research__research_analyze_known_binders` — SAbDab antibody database `mcp__by-research__research_find_similar_targets` — homologs with known binders

04
Phase 4: TRIANGULATE

Cross-validate findings across sources: Do structural data (PDB interface residues) match literature claims about binding sites? Do known antibody epitopes from SAbDab agree with PubMed-reported epitopes? Are binding affinity values consistent across publications? Assign confidence to each finding:

05
Phase 5: SYNTHESIZE

Structure findings into a coherent narrative covering: Target overview — function, structure, therapeutic relevance Known binders — existing antibodies/nanobodies from SAbDab + literature Epitope analysis — consensus hotspot residues, interface character Design strategy — recommended modality, scaffolds, campaign tier

06
Phase 6: CRITIQUE

Challenge the synthesis using three distinct review personas. Each persona applies a different lens to the research. The critique output must label every concern with the persona that raised it (e.g., `[Skeptical Practitioner]`, `[Adversarial Reviewer]`, `[Implementation Engineer]`). Persona 1: Skeptical Practitioner

07
Phase 7: REFINE

Address critique findings: Run additional targeted searches to fill gaps Strengthen weak claims with additional sources Explicitly note remaining uncertainties (do not hide them) Update confidence levels in `validated_findings.json`

08
Phase 8: PACKAGE

Write final outputs to the campaign directory: `research/research.md` — formatted report (see Output Format below) `research/recommended_hotspots.json` — residue list for the design agent `research/design_recommendation.json` — modality, scaffolds, tier, parameters Update campaign state with research summary

Discussion

Powered by GitHub Discussions. Sign in with GitHub to comment, react, or subscribe.