agentleFS
tatashbartatashbar/Economic-Modeling-Studio/.agents/skills/country-typology/SKILL.md

country-typology

Skill4 starsChanged 6 days agoMITView on GitHub

Groups countries into policy-relevant types from data, by transparent rules or clustering, and maps each type to tailored policy options.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 5 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If a required input is ABSENT, that part is NOT ASSESSED — NO DATA: do not estimate or invent it.

    line 27
  2. Very specific:

    Collaboration: Follow .claude/docs/collaboration-protocol.md: ask questions first, present options, get approval before writing files, and …

    line 29
  3. Distinctive:

    After the rules: confirm types and indicators

    line 56
  4. Distinctive:

    After the assignment: review the country lists and borderline cases

    line 57
  5. Distinctive:

    Before the policy table: confirm the source of each recommendation

    line 58
  6. Distinctive:

    Every country assignment can be traced to data and a rule

    line 72
  7. Distinctive:

    Every policy response has a stated source or rationale

    line 74

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md
<!-- Generated from .claude/skills/country-typology/SKILL.md by tools/sync_agents.py. Edit the source, then run the script. -->

/country-typology — Country Typology

Category: Analysis · Primary agent: cross-country-indicators-specialist

Groups countries into policy-relevant types from data, by transparent rules or clustering, and maps each type to tailored policy options.

Arguments: whatever the user wrote after the command name.

Purpose

Turn general advice into tailored advice: assign each country to a type (for example high debt with large refinancing needs, commodity exporter, large exposure to state-owned enterprises and guarantees, lower capacity, moderate debt with high shock exposure) using indicators and transparent rules or clustering, list the countries in each type, and map each type to its main issue and priority policy responses. When a user brings a typology table, the skill implements it as rules on data, shows which countries fall in each type, and tests how stable the assignment is.

Use it when recommendations must differ by country circumstance, or a proposed typology needs data behind it.

Do not use it when the work is a single country's policy options (/policy-simulation).

Before you start

  1. Check required inputs. For each input below record FOUND or ABSENT (never 'assumed'):
    • Indicator panel for the countries (debt, financing needs, commodity dependence, contingent liabilities, revenue capacity, buffers, shock exposure)
    • Country classification (export structure, income, region)
    • A proposed typology or policy table (optional) If a required input is ABSENT, that part is NOT ASSESSED — NO DATA: do not estimate or invent it; name what is missing and which skill produces it. If all required inputs are absent, stop and say so.
  2. Check confidentiality. Identify the data classification level of the inputs (.claude/docs/confidentiality-and-ai-use.md). Do not proceed with restricted or confidential data in a way the policy does not allow.
  3. Collaboration. Follow .claude/docs/collaboration-protocol.md: ask questions first, present options, get approval before writing files, and update production/session-state/active.md after each approved section.

Agents

  • cross-country-indicators-specialist — primary agent; coordinates the work and integrates results
  • debt-sustainability-analyst — Builds debt sustainability analyses
  • fiscal-risk-specialist — Assesses fiscal risks
  • policy-relevance-specialist — Clarifies the decision, audience and deadline, and reviews outputs for decision-relevance
  • policy-brief-writer — Writes policy briefs

Each agent is defined in .claude/agents/<name>.md. If your tool runs subagents, run these agents as subagents, in parallel where the steps are independent; otherwise read each agent's file and apply its instructions in turn. Subagents only write to the paths named in your instructions.

Key design decisions (present options, let the user choose)

DecisionTypical optionsWhat it changes
Methodrules on indicator thresholds; clustering (k-means, hierarchical) on standardized indicators; expert assignment checked against dataTransparency versus data-driven grouping
Overlapone type per country; several types allowed with a primary typeSimplicity versus realism
Indicatorsone or two per type; a composite per typeRobustness and explainability
Policy mappinga supplied table; literature-based options; bothSource of the recommendations

Workflow

  1. Agree the types, the indicators and rule (or clustering) behind each, and whether countries may belong to several types
  2. Build the indicator panel; assign countries; list the countries in each type with the values that put them there
  3. Stability: change thresholds, years and indicators; report countries that switch type
  4. Map each type to its main issue and priority responses, with the evidence or source for each response
  5. Sense-check with the user: countries whose assignment looks wrong are examined, not silently moved

Checkpoints with the user

  • After the rules: confirm types and indicators
  • After the assignment: review the country lists and borderline cases
  • Before the policy table: confirm the source of each recommendation

Common pitfalls

  • Types defined so loosely that every country qualifies
  • Recommendations attached to types without evidence or source
  • Clusters with no economic interpretation
  • Hiding borderline countries that switch type under small changes

Output

docs/models/<name>/country-typology.md; code and data in models/econometrics/<name>/

The report lists the types, rules or clustering method, the countries in each type with their indicator values, the stability results and the policy table by type. Every run saves its results to output/<run-id>/ in the model folder, with a run log (.claude/rules/model-code.md).

Gate / verdict

  • Every country assignment can be traced to data and a rule
  • Borderline and switching countries are listed
  • Every policy response has a stated source or rationale

Suggested next steps

  • /policy-brief
  • /scenario-analysis

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.