Skip to main content
skill CX Codex

grill-mel — Claude Code skill: what it does, how to install

grill-mel — a skill that stress-tests MEL designs before drafting: theories of change, indicator sets, results frameworks. Asks one question at a time, recommends an answer, and pushes back on every untested assumption.

no license 491 installs ★ 3 updated 2026-10-23

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add gasserane/personal-skills --skill grill-mel -a claude-code
Open on GitHub

What grill-mel does

grill-mel interviews you on a MEL design: theory of change, evaluation plan, indicator set, or results framework. Each question comes with a recommendation — the skill does not leave you staring at a blank page, it proposes a position and waits for pushback. It walks the branches: tested versus untested assumptions, OECD-DAC criteria, output versus outcome versus impact, disaggregation by age, gender, disability and geography, data gaps. If a fact is not in scope, it flags and asks, never invents.

It stops when the branches are resolved. Hands the sharpened design off to /toc-builder, /indicator-designer, or the full /ann workflow.

Built for MEL (monitoring, evaluation, learning) and SRHR (sexual and reproductive health and rights) projects. Outside those domains it has no use.

What it helps with

  • test a theory of change before writing the document
  • distinguish output from outcome in a results framework
  • spot untested assumptions in an evaluation design
  • check whether indicators are disaggregated by age, gender, disability and geography
  • figure out which OECD-DAC criteria the design actually covers

Example prompts for the agent

  • Test my theory of change
  • Stress-test this indicator set before I submit
  • Walk me through the OECD-DAC criteria for this results framework

How to install grill-mel

1Skills CLI (Vercel)

npx skills add gasserane/personal-skills --skill grill-mel -a claude-code

2Manually: put the files in place

~/.claude/skills/grill-mel/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/grill-mel/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about grill-mel

How to install grill-mel in Claude Code?

Run npx skills add gasserane/personal-skills --skill grill-mel -a claude-code, the file will be in ~/.claude/skills/grill-mel/SKILL.md.

What does grill-mel do?

Asks questions about a MEL design one at a time, recommends an answer for each, and pushes back on untested assumptions. Tests theories of change, indicators, and results frameworks before drafting starts.

How is grill-mel different from the generic grilling skill?

grill-mel knows MEL and SRHR branches: OECD-DAC criteria, indicator disaggregation, output versus outcome distinction. The generic grilling handles any plan, but lacks that specificity.

Similar skills

All skills for Writing and docs →

From the same set44

The whole set →
improve-system Mines session history and QA logs for recurring failure patterns, then proposes targeted system edits. Runs on demand via /improve-system or fires automatically on a 14-day cadence. CX Codex +1 581installs accreditation-desk-review Conducts IPPF Member Association accreditation desk reviews against Cycle 4 ARV guidelines. Evaluates submitted evidence and records findings in a Word report. CX Codex +1 581installs implementation-pack Builds a post-award project-management pack from a proposal and grant terms. Run it to set up project tracking or use /implementation-pack --donor cerv. CX Codex +1 576installs learned learned runs a four-question post-project review and logs verbatim takeaways into Obsidian. Useful for capturing practical lessons right after delivering work. CX Codex +1 555installs research-scoping The research-scoping skill for Claude Code maps research questions and tensions in one pass before heavy workflows, helping decide whether to escalate. CX Codex +1 438installs decision-debate decision-debate runs a structured debate between autonomous specialist agents on tough MEL and SRHR design choices. It helps review clashing perspectives before committing to a plan. CX Codex +1 383installs check-deliverable check-deliverable — a skill for Claude Code invokes the qa-reviewer subagent to audit a single finished text deliverable. It validates the draft against quality gates without triggering the full Ann workflow. CX Codex +1 373installs mel-discipline mel-discipline makes the agent walk five stages when writing an analytical deliverable, from scoping to an honest report on what it checked. Useful when a deadline pushes you to cite a source you never opened. CX Codex +1 355installs

Project sponsors

Companies that keep this analytics open