M&E Review
Professional monitoring, evaluation, accountability and learning (MEAL) skills for Claude.
Paste a logframe, theory of change, M&E plan, survey, evaluation TOR, sampling plan, baseline report, progress report, proposal, or indicator list and get a structured quality review in seconds.
Built by M&E Studio.
New in 1.5.0: the me-workspace plugin adds twelve job skills and three workflow packs, and each skill description now says which skill owns which job. The plugin ships as its own marketplace, so installation is two commands (see below). Each review is a single self-contained skill.
New in 1.3.0: six additional reviews covering theories of change, proposal M&E sections, sampling plans, data quality assessments, baseline reports and progress reports. Twelve reviews in total.
What It Does
| Skill | Use It When You Have... |
|---|---|
/me-review:logframe-review |
A logframe or results framework to check for logic gaps, weak indicators and missing assumptions |
/me-review:theory-of-change-review |
A theory of change, causal pathway or results chain to check for logical coherence and assumption quality |
/me-review:indicator-quality |
A list of indicators to score against SMART and CREAM criteria |
/me-review:me-plan-review |
An M&E plan, MEAL plan, PMP or MEL plan to assess for completeness and feasibility |
/me-review:proposal-me-review |
The M&E section of a proposal, grant application or concept note |
/me-review:tor-review |
An evaluation TOR or SOW to review for methodological rigor, scope and budget realism |
/me-review:survey-review |
A survey, questionnaire, KII guide or FGD guide to check for design quality and ethics |
/me-review:sampling-plan-review |
A sampling plan, sample size calculation or sampling frame to check for method appropriateness |
/me-review:baseline-report-review |
A baseline report or pre-intervention study to check for indicator coverage and target validation |
/me-review:progress-report-review |
A quarterly, annual or donor progress report to check for data plausibility and variance explanation |
/me-review:data-quality-review |
A DQA report, data quality audit or verification report to check for VIPRT coverage and rigor |
/me-review:evaluation-report-review |
A final evaluation report to assess for quality, evidence anchoring and limitations |
Each skill produces a scored review with section-by-section ratings (PASS / PARTIAL / FAIL), a summary verdict, and prioritized recommendations.
Installation
Two commands. This repository is its own plugin marketplace, so you add it once and then install from it.
/plugin marketplace add mestudiohq/me-review
/plugin install me-review@me-review
Run these inside Claude Code. If the install summary says Run /reload-plugins to activate., run that too.
To check it worked, type /me-review: and you should see all twelve reviews listed.
Claude Desktop users can install the same plugin through the plugin browser in Settings.
Companion Plugin: me-workspace (free)
The same marketplace holds me-workspace, for Claude Code. It reads a whole project folder: 11 checks compare your documents against each other (TOR against logframe, M&E plan coherence, progress-report coverage), and 8 creates draft new M&E documents with source tags. Install it after adding the marketplace:
/plugin install me-workspace@me-review
Details: plugins/me-workspace.
How to Use
Type a slash command, then paste your document:
Logframe review:
/me-review:logframe-review
Objective: Improved health outcomes for rural women
Output 1: 500 health workers trained
Output 2: 12 health facilities upgraded
Indicator: Number of women accessing services (target: 10,000)
Theory of change review:
/me-review:theory-of-change-review
[paste your theory of change, causal pathway, or results chain here]
Indicator quality check:
/me-review:indicator-quality
1. Number of health workers trained (target: 500)
2. % of facilities with adequate supplies (baseline: 42%)
3. Women's empowerment improved
Evaluation TOR review:
/me-review:tor-review
[paste your evaluation TOR here]
M&E plan review:
/me-review:me-plan-review
[paste your M&E plan, MEL plan, PMP, or MEAL plan here]
Sampling plan review:
/me-review:sampling-plan-review
Population: 12,400 households across 3 districts
Method: Two-stage cluster sampling, 40 clusters of 15 households
Sample size: 600, calculated for 5% margin of error at 95% confidence
Disaggregation required: sex of household head, district
Evaluation report review:
/me-review:evaluation-report-review
FINDINGS
Finding 1: Student learning improved 23% on average (treatment: 65%, control: 53%)
- Evidence: Pre/post learning assessments in 60 schools, 4,800 students
- Limitations: Comparison group non-randomly selected; causality estimates may reflect contextual differences
Finding 2: Teacher practice improved significantly (78% competency-based teaching in treatment vs. 12% in control)
- Evidence: Classroom observations using validated rubric, 240 observations across sites
- Limitation: Self-selection of trained teachers may bias estimates
RECOMMENDATIONS
Priority 1: Scale to 200 additional schools by 2025 ($1.86M investment)
Priority 2: Establish peer learning networks to sustain practice changes
Priority 3: Conduct 2-year follow-up to verify learning gains persist
Survey review:
/me-review:survey-review
[paste your survey, questionnaire, or KII guide here]
Claude will review the document and return a structured report.
You can also just paste a document without typing a command. Claude will detect what it is and apply the right review automatically.
More free M&E resources
This plugin is one part of a free toolkit for M&E practitioners.
| Resource | What it is |
|---|---|
| AI prompt library | Task-specific prompts for M&E work, in English, Spanish and French |
| Free M&E tools | Sampling calculator, DQA scorecard, SMART indicator checker, method selector, evaluation readiness check |
| M&E reference library | Plain-language entries on M&E methods, terms and comparisons |
| Method guides | Step-through guides for planning and running evaluations |
| Plugin page | Installation help and what each skill checks |
License
MIT License. Copyright (c) 2026 MEStudio