Proposal_Judge (Proposal Pre-Review)

An internal pre-reviewer that stress-tests a research proposal to remove scoring, clarification, and disqualification risks.

Overview
Version
v1.0.0
Created
2025-12-16
Updated
2025-12-16
proposal-reviewrfpmarket-researchdue-diligencequality-control
proposal-judgeprejudge
Key functions
  • Extracts (or explicitly assumes) RFP requirements and flags fit gaps or misinterpretations
  • Detects methodology/logic/differentiation issues that may look templated, reused, or low-effort
  • Checks execution credibility: sampling, quality control, timeline, staffing, and numeric consistency
  • Formats every finding as Issue/Risk/Fix Direction with actionable remediation guidance
  • Anticipates likely evaluator questions to support pre-submission Q&A readiness
Technical details
_id
g-693e676668108191b120ebe6c04d17d2
gpt_id
g-693e676668108191b120ebe6c04d17d2
viz1
public
viz2
show_url
language
en
Other fields
additional_features
["Output is locked to 4 blocks (summary / layered findings / top 5 fixes / likely questions)", "Treats ambiguity as a problem and prioritizes worst-case evaluator interpretations"]
example_commands
["Paste the proposal text and output only: Summary / Fit/Substance/Execution findings / Top 5 fixes / Likely questions.", "Here is the RFP and a proposal outline—identify the Top 5 high-risk fit gaps and how to fix them.", "Review only the sampling, QC, and schedule sections; produce 5 evaluator questions we should prepare for."]
gpt_id
g-693e676668108191b120ebe6c04d17d2
ideal_use_cases
["Final pre-submission proposal audit (missing items, contradictions, unclear claims, trust-reducing wording)", "Risk identification using an 'assumed RFP' when requirements or internal inputs are incomplete", "Pre-building evaluator Q&A around sampling, methods, schedule, deliverables, and differentiation"]
limitations
["Does not declare Go/No-Go; does not rewrite or ghostwrite proposal sentences", "Cannot verify facts beyond provided materials (e.g., internal rules or latest notices) without evidence", "If a specific scoring rubric exists, an additional review aligned to that rubric is recommended"]
target_users
["Research proposal authors / PMs", "Proposal QA/QC reviewers", "Research & consulting project leads"]