← Back to Skills Marketplace
1065
Downloads
3
Stars
1
Active Installs
19
Versions
Install in OpenClaw
/install nonprofit-rbm-logic-model
Description
Turn incomplete nonprofit project inputs into a donor-aligned proposal package, a transparent evidence trail, and a defensible Go / Conditional Go / No-Go su...
Usage Guidance
This instruction-only skill appears coherent and does not request credentials or installs. Before using it, provide the donor call text and supporting evidence (budgets, partner commitments, baseline data) rather than pasting sensitive PII; the skill explicitly refuses to invent or assert undocumented facts and will mark unverified items. If you need the agent to retrieve donor guidelines or documents from the web, do not expect the skill to do that autonomously — supply those sources directly. Overall this looks safe to install, but avoid pasting confidential credentials or personally-identifiable data into prompts.
Capability Analysis
Type: OpenClaw Skill
Name: nonprofit-rbm-logic-model
Version: 1.1.2
The skill bundle 'nonprofit-rbm-logic-model' is a set of markdown-based instructions designed to guide an AI agent in evaluating and drafting nonprofit proposals. It emphasizes evidence-based decision-making, transparency, and donor alignment, explicitly forbidding the fabrication of data or the masking of structural weaknesses. There is no executable code, no network activity, and no evidence of malicious intent or prompt injection; the instructions are focused on professional rigor and risk assessment.
Capability Assessment
Purpose & Capability
The name and description (go/no-go decision support for nonprofit proposals) match the SKILL.md content. The skill is instruction-only and does not request unrelated credentials, binaries, or config paths.
Instruction Scope
SKILL.md is prescriptive about inputs to collect, evidence discipline, and output modes. It does not instruct the agent to read system files, environment variables, or fetch remote resources on its own; it expects user-supplied donor text/evidence. The rules against inventing data and the Missing-input rule are explicit and appropriate for the purpose.
Install Mechanism
No install spec and no code files — the skill is instruction-only, which minimizes on-disk risk. There are no downloads or third-party packages referenced.
Credentials
The skill declares no required environment variables, credentials, or config paths. Its functioning (prompt-based collection of proposal inputs and review rules) does not require elevated access or secrets.
Persistence & Privilege
Flags show always:false and default autonomous invocation; nothing in the skill requests permanent presence or modification of other skills or system-wide settings.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install nonprofit-rbm-logic-model - After installation, invoke the skill by name or use
/nonprofit-rbm-logic-model - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.2
- Skill description and instructions extensively rewritten for clarity and brevity.
- Added a required “Mandatory opening block” summarizing core submission context at the start of every output.
- Streamlined and clarified input requirements (minimum input contract) and missing-input handling rules.
- Enhanced evidence discipline section, requiring explicit labeling of facts, assumptions, hypotheses, and unknowns.
- Refined, reorganized, and further explained the workflow, output structure, and mode system for greater transparency and stronger submission discipline.
v1.1.1
- Renamed the skill to "nonprofit-proposal-go-no-go-engine" and updated documentation to reflect this, emphasizing a decision-first proposal review approach.
- Simplified required files by removing legacy SKILL.md and _meta.json, consolidating all documentation into Skill.md.
- Refined audience and use cases; clarified what the skill does and does not support, including stricter refusal to fabricate evidence or hide weaknesses.
- Streamlined mode definitions and workflow steps for concept, LOI, full proposal, review, donor-fit, and express outputs.
- Improved clarity in minimum input requirements and missing input handling, with clear fallback behavior for incomplete user inputs.
v1.1.0
Major rewrite: tighter positioning, mode-based workflow, strict input/output contracts, stronger evidence guardrails, and submission decision gating (Go/Conditional/No-Go).
v1.0.15
Strengthened execution playbook: clearer best-fit/not-for boundaries, 60-second preflight, tighter evidence policy language, and stronger submission gate framing (Go/Conditional/No-Go).
v1.0.14
Refine summary for donor-ready outcomes, submission readiness, and trigger coverage.
v1.0.13
Clarify invocation patterns as pseudo-commands, not local binary requirement; enforce strict citation metadata; add source-limited fallback to prevent fabricated evidence
v1.0.12
Strengthen donor-grade rigor: evidence note, unit-cost assumptions, compliance gates, verified vs placeholder baselines, 2-week validation sprint.
v1.0.11
- Renamed skill from nonprofit-rbm-logic-model to nonprofit-impact-orchestra for clarity and broader relevance.
- Improved and condensed instructions for easier onboarding and quick start.
- Clarified supported use cases, outputs, and workflow steps.
- Updated intake template fields for more precise scoping (e.g., explicitly added "problem" and "need now" fields).
- Added and clarified core rules: realism, structured outputs, transparency about claim confidence, minimum questioning, and implementation focus.
- Expanded quick reference, modes, and output delivery options, including JSON export and "donor-fit" review mode.
v1.0.10
- Clarified and streamlined documentation for ease of use, reorganizing step-by-step workflow instructions.
- Updated audience guidance for each output mode and improved descriptions (e.g., donor/board/community register options).
- Made free-form input handling and clarification logic explicit.
- Added a new "Compliance Score" step summarizing readiness with checklist outputs.
- Improved instructions for JSON-ready and Excel-friendly outputs.
- Catalogued all steps, including donor adaptation, human impact narrative, and natural language polish, in a clear, concise structure.
v1.0.9
**Major upgrade: Now supports multiple modes, more compliance features, and modular workflow customization.**
- Adds one-command modes: Express, CFP extraction, Concept Note, LOI, Review, Peer Review, Compare, and modular entry from any step.
- Expands coverage to GESI (Gender Equality & Social Inclusion) analysis, Do No Harm/safeguarding, and explicit donor compliance scoring.
- Refines intake: use structured templates, free-form prompts, or auto-extraction from pasted Calls for Proposals.
- Introduces multi-audience register (donor, board, community) and multilingual output.
- All outputs now include confidence flags for statistics and web-source checks (where enabled).
- Improved budget tools, risk logic, scenario planning, and proposal adaptation to specific donor criteria.
v1.0.8
**Major upgrade: Now an end-to-end orchestrator for creating complete, polished nonprofit project packages.**
- Fully redesigned as Nonprofit Impact Orchestra — a multi-step workflow that turns a short project idea into a structured, donor-ready grant proposal.
- Adds deep strategic context analysis (including PESTLE, stakeholder mapping, horizon scanning).
- Builds out all key logic frameworks: RBM, Theory of Change, logframe matrix, SDG mapping.
- Includes detailed budget breakdown, risk & scenario assessment, donor-specific proposal tweaks, and human impact narrative.
- Features interactive checkpoints for review, editing, and customization at each key stage.
- Final output is professionally formatted for external use, with structured sections, export blocks, and Excel-friendly tables.
v1.0.7
Skill renamed and refocused with streamlined documentation.
- Renamed from "nonprofit-rbm-logic-model" to "ngo-program-design-suite"
- Updated description and tags for broader NGO program design and grant proposal use
- Completely rewritten documentation with simplified instructions and donor-oriented structure
- Removed references to internal scripts, reference files, and gating system
- Five files (including LICENSE and implementation scripts) were removed for a cleaner, documentation-only release
v1.0.6
Product hardening: added deterministic RBM quality gate script, gate status requirement in output, and evidence-first quality standard for machine-readable deliverables.
v1.0.5
Improve conversion and clarity: tighter value-focused description, cleaner workflow, stricter output schema, and stronger quality standards.
v1.0.4
Minor updates
v1.0.3
Expand abbreviations in skill description and prompts for clarity (RBM, SMART, SDG, NGO, USAID, UN, EU)
v1.0.2
Improve skill description for discoverability and audience clarity
v1.0.1
Add Theory of Change as required first output section and renumber output format
v1.0.0
Initial publish with 5-level RBM chain, SMART indicators, SDG alignment, and monitoring guidance
Metadata
Frequently Asked Questions
What is Nonprofit Proposal Decision Engine?
Turn incomplete nonprofit project inputs into a donor-aligned proposal package, a transparent evidence trail, and a defensible Go / Conditional Go / No-Go su... It is an AI Agent Skill for Claude Code / OpenClaw, with 1065 downloads so far.
How do I install Nonprofit Proposal Decision Engine?
Run "/install nonprofit-rbm-logic-model" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Nonprofit Proposal Decision Engine free?
Yes, Nonprofit Proposal Decision Engine is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Nonprofit Proposal Decision Engine support?
Nonprofit Proposal Decision Engine is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Nonprofit Proposal Decision Engine?
It is built and maintained by Vasiliy (@vassiliylakhonin); the current version is v1.1.2.
More Skills