Status: Active
Date: 2026-06-14
Author: Cytognosis Foundation
Audience: researchers, grant writers, communications leads
Tags: quick-reference, cytognosis-writer
cytognosis-writer — Quick Reference
One line: Load this skill for grants, proposals, pitch decks, science narratives, and any external-facing Cytognosis content; always read the funder-specific reference file before drafting.
Full doc: cytognosis-writer.md
Key Concepts
| Term |
Definition |
| Revelation Arc |
Cytognosis narrative structure: Mystery (diagnostic failure) → Insight (cellular intelligence) → Resolution (disease interception). Applies to all external writing. |
| Three Blindspots |
The three gaps in current healthcare that Cytognosis addresses. Detail in references/science-platform.md. |
| GPS for Human Health |
The core Cytognosis metaphor: a health navigation system that detects disease early. |
| Bridge line |
Rhetorical anchor: "Cytognosis exists so no one else waits decades for answers." |
| Language blocks |
Pre-approved, reusable text for grants. Never draft from scratch; assemble from blocks in references/grant-strategy.md. |
| FAIR |
Findable, Accessible, Interoperable, Reusable. Never assert FAIR compliance without mapping each principle to infrastructure. |
Decision Tree
| Task |
Read |
| Writing any grant or proposal |
references/science-platform.md + references/grant-strategy.md |
| Writing about open science, FAIR, or DMP |
references/openness-policy.md |
| Extracting grant content from meeting transcript |
references/meeting-extraction.md |
| Writing about Cytognosis science or technology |
references/science-platform.md |
| NIH-specific writing |
references/agencies/nih.md |
| ARPA-H-specific writing |
references/agencies/arpa-h.md |
Strategic Terminology
| Use |
Instead of |
| intercept |
prevent |
| detect and intercept |
diagnose and treat |
| years before symptoms |
early detection |
| people / individuals |
patients (except clinical context) |
Tone by Audience
| Audience |
Register |
| ARPA-H |
Bold, moonshot, commercial transition |
| NIH |
Rigorous, conservative, reproducibility-focused |
| FRO |
Infrastructure, engineering, public utility |
| NSF |
Scientific merit, broader impacts |
| Public / social |
Warm, accessible |
| Website |
Clear, confident |
| Scientific papers |
Precise, rigorous |
Voice Rules (Abbreviated)
| Rule |
Correct |
| Em dashes |
Forbidden. Use commas or semicolons. |
| Voice |
Active present. "We detect disease." |
| "Patients" |
Only in clinical context. |
| Hype words |
No "revolutionary", "cure", "breakthrough", "disrupt" |
| Intercept vs. prevent |
Always "intercept" |
| Capitalization |
Cytoverse, Cytoscope, Cytonome, Helix model (always capital) |
Common Patterns
# NIH grant section workflow
1. READ references/science-platform.md
2. READ references/agencies/nih.md
3. READ references/grant-strategy.md (language blocks)
4. Apply Revelation Arc: Mystery → Insight → Resolution
5. Save as: Grant-NIH-R01-<Section>-v1-<YYYY-MM>.md in 01-Grants-and-Funding
# Meeting-to-grant extraction
1. READ references/meeting-extraction.md
2. Apply extraction personas
3. Output: structured claims, evidence statements, milestone language
4. Cross-reference against grant-strategy.md language blocks
# FAIR compliance statement
1. READ references/openness-policy.md
2. Map each FAIR principle to Cytognosis infrastructure
3. Never write "We follow FAIR" without the mapping table
Reference Files
| File |
When to Read |
references/science-platform.md |
Platform narrative, Three Blindspots, founder story |
references/grant-strategy.md |
Funder-specific strategy, language blocks |
references/openness-policy.md |
Licensing, FAIR, open AI models, open-core rule |
references/meeting-extraction.md |
Extracting grant content from transcripts |
references/agencies/ |
Per-funder guidelines (ARPA-H, NIH, NSF, FRO, DARPA, DOE, Wellcome) |
NEVER
- NEVER draft grant sections from scratch; assemble from language blocks in
grant-strategy.md
- NEVER guess funder requirements; read their
references/agencies/<funder>.md
- NEVER write "We follow FAIR" without a per-principle infrastructure mapping
- NEVER use em dashes, hype words, or passive voice
- NEVER suggest GPL or copyleft licensing
- NEVER ask which sections to draft; draft all requested sections
Error Quick-Fix
| Error / Symptom |
Fix |
| Grant section uses wrong register |
Check tone table above; NIH = rigorous/conservative, ARPA-H = bold/moonshot |
| "Revolutionary" or "breakthrough" in draft |
Replace with specific evidence: "X% improvement", "first platform to..." |
| Missing FAIR mapping |
Open references/openness-policy.md and map each F/A/I/R to actual infrastructure |
| Writing from scratch instead of language blocks |
Open references/grant-strategy.md; locate the relevant block; adapt |
See Also