agentleFS
tatashbartatashbar/Economic-Modeling-Studio/.agents/skills/build-io-sam/SKILL.md

build-io-sam

Skill4 starsChanged 6 days agoMITView on GitHub

Builds input-output, supply-use and SAM frameworks with multipliers, linkages, multi-regional and environmentally extended extensions.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 6 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    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:

    Document concordances between classifications

    line 51
  4. Distinctive:

    Run impact analyses and state the fixed-coefficient limitations alongside results

    line 56
  5. Distinctive:

    Document concordances, data sources, adjustments and reproduction steps

    line 58
  6. Distinctive:

    After choosing framework and classification: confirm before construction

    line 62
  7. Distinctive:

    After balancing: confirm the adjustment sizes are acceptable

    line 63
  8. Distinctive:

    Before impact runs: confirm the shock definition

    line 64

≈ “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/build-io-sam/SKILL.md by tools/sync_agents.py. Edit the source, then run the script. -->

/build-io-sam — Build IO SAM

Category: Build · Primary agent: input-output-sam-specialist

Builds input-output, supply-use and SAM frameworks with multipliers, linkages, multi-regional and environmentally extended extensions.

Arguments: whatever the user wrote after the command name.

Purpose

Build an input-output, supply-use or social accounting framework that is balanced, documented and ready for multiplier and impact analysis, or as the benchmark dataset for a CGE model. The output is a reproducible table with automated balance checks, multipliers and linkages, and a clear statement of the fixed-coefficient limits that apply to any impact result.

Use it when the question concerns sectoral linkages, demand-driven multipliers, value-added content of trade, environmental footprints, or when a CGE model needs a benchmark SAM.

Do not use it when prices, substitution and supply constraints matter for the answer (/build-cge) or when the shock propagates mainly through supply bottlenecks (/supply-chain-shock).

Before you start

  1. Check required inputs. For each input below record FOUND or ABSENT (never 'assumed'):
    • Supply-use or input-output tables
    • National accounts for the base year
    • Classification concordances 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

  • input-output-sam-specialist — primary agent; coordinates the work and integrates results
  • national-accounts-specialist — Specializes in national accounts and macro statistics
  • production-networks-specialist — Models production networks and supply chains
  • trade-gvc-specialist — Models trade and global value chains
  • carbon-pricing-emissions-specialist — Analyzes carbon pricing and emissions policy

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
Frameworknational IO, supply-use, SAM, multi-regional IOQuestions answerable; data needs
Technology assumptionindustry, productSymmetric table construction
Aggregationnumber and grouping of sectorsAggregation bias; readability
Balancing methodRAS, GRAS, cross-entropyTreatment of negatives; adjustment size
Extensionsvalue added in trade, emissions, employmentSatellite outputs

Workflow

  1. Choose the framework (national IO, supply-use, SAM, multi-regional) and the sector classification; ask the user to approve
    • Document concordances between classifications
  2. Construct and balance the tables (RAS/GRAS/cross-entropy where needed); automate balance checks
    • Report the size of each adjustment
  3. Compute Leontief/Ghosh inverses, multipliers, linkages and, if relevant, value-added and environmental extensions
    • Distinguish Type I and Type II multipliers; flag Ghosh interpretation limits
  4. Run impact analyses and state the fixed-coefficient limitations alongside results
    • Separate direct, indirect and induced effects; domestic and imported content
  5. Document concordances, data sources, adjustments and reproduction steps
  6. Register the model as a lego set (/lego-map; a stand-alone table uses the core-io-leontief baseplate): a baseplate plus bricks with declared studs (variables, units, timing) and stubs, so components can later be attached, detached or swapped

Checkpoints with the user

  • After choosing framework and classification: confirm before construction
  • After balancing: confirm the adjustment sizes are acceptable
  • Before impact runs: confirm the shock definition

Common pitfalls

  • Gross output multipliers reported as value added or welfare
  • Summing multipliers across sectors
  • Using an outdated table for a changed economy
  • Ignoring imports, which overstates domestic effects

Output

models/io-sam/<name>/ (+ model.lego.json)

The folder holds source data, construction and balancing scripts, the balanced table, multiplier and linkage tables, impact-analysis scripts and results, the documentation note and the lego manifest. Every run saves its results to output/<run-id>/ in the model folder, with a run log (.claude/rules/model-code.md).

Gate / verdict

  • Row and column totals balance to tolerance
  • Hawkins-Simon condition verified
  • Adjustments and concordances documented
  • Lego manifest passes python tools/lego.py check
  • Fixed-coefficient limitations are stated with every impact result
  • Impact results separate domestic and imported effects

Suggested next steps

  • /build-cge
  • /supply-chain-shock
  • /validate-model
  • /lego-map

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.