agentleFS
tatashbartatashbar/Economic-Modeling-Studio/.agents/skills/build-semi-structural/SKILL.md

build-semi-structural

Skill4 starsChanged 6 days agoMITView on GitHub

Builds a small or medium semi-structural gap/projection model with calibration, filters and baseline.

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 26
  2. Very specific:

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

    line 28
  3. Distinctive:

    Create the baseline run and shock-response pack

    line 52
  4. Distinctive:

    Write tests: steady-state residuals, accounting identities, IRF sign checks

    line 54
  5. Distinctive:

    Document and hand over to validation

    line 55
  6. Distinctive:

    After the equations: confirm blocks and the policy rule

    line 60
  7. Distinctive:

    After calibration: confirm IRF signs and magnitudes make sense

    line 61
  8. Distinctive:

    After the baseline: confirm the filtered gaps and trends tell a plausible story

    line 62

≈ “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-semi-structural/SKILL.md by tools/sync_agents.py. Edit the source, then run the script. -->

/build-semi-structural — Build Semi Structural

Category: Build · Primary agent: semi-structural-gap-modeler

Builds a small or medium semi-structural gap/projection model with calibration, filters and baseline.

Arguments: whatever the user wrote after the command name.

Purpose

Build a small or medium semi-structural gap and projection model: aggregate demand, Phillips curve, policy rule, exchange-rate and external blocks, with unobserved trends and gaps filtered from data. The output is a transparent, calibrated model with a baseline, a shock-response pack and tests, suitable for forecasting and policy analysis in an FPAS.

Use it when a transparent projection model is needed for regular forecasting and policy discussions, data are limited, or counterparts need a model they can own and explain.

Do not use it when microfoundations, welfare or regime changes are central (/build-dsge), or an accounting framework is needed (/build-financial-programming).

Before you start

  1. Check required inputs. For each input below record FOUND or ABSENT (never 'assumed'):
    • Approved model-spec
    • Data sheet 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

  • semi-structural-gap-modeler — primary agent; coordinates the work and integrates results
  • state-space-kalman-specialist — Builds state-space models
  • monetary-policy-rules-specialist — Designs and evaluates monetary policy rules
  • expectations-learning-specialist — Models expectations formation

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
Sizecore (4 equations), extended (fiscal, external, financial blocks)Story richness versus maintenance
Trendsrandom walks, local linear trends, steady-state anchorsPotential output and neutral rate paths
Policy ruleinflation-forecast-based, Taylor-type, exchange-rate termsPolicy experiments
Exchange rateUIP with premium, managed regimeExternal transmission
Parameter choicecalibration, Bayesian estimation, mixFit versus economic discipline

Workflow

  1. Implement the equations and gap filters; set up code in models/semi-structural/<name>/
    • Separate files for parameters, equations, filtering and simulation
  2. Calibrate or estimate; check steady state and impulse responses against expected signs
    • Compare persistence and pass-through with empirical evidence
  3. Create the baseline run and shock-response pack
    • Standard shocks: demand, cost-push, policy, exchange-rate premium, foreign demand, commodity prices
  4. Write tests: steady-state residuals, accounting identities, IRF sign checks
  5. Document and hand over to validation
    • Model card and user guide (/model-card)
  6. Register the model as a lego set (/lego-map): 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 the equations: confirm blocks and the policy rule
  • After calibration: confirm IRF signs and magnitudes make sense
  • After the baseline: confirm the filtered gaps and trends tell a plausible story

Common pitfalls

  • Filtered gaps driven by end-point problems
  • Parameters tuned to fit without economic plausibility
  • Exchange-rate blocks that ignore the actual regime
  • Overlays used to fix a misspecified equation

Output

models/semi-structural/<name>/ (+ model.lego.json)

The folder holds model code, calibration, filtering and baseline files, the shock-response pack, tests, documentation 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

  • Steady state residuals ≈ 0
  • IRF signs and magnitudes are plausible and documented
  • Lego manifest passes python tools/lego.py check
  • Filtered trends and gaps are reviewed for end-point sensitivity
  • Tests for identities and IRF signs pass

Suggested next steps

  • /validate-model
  • /forecast-round
  • /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.