agentleFS
tatashbartatashbar/Economic-Modeling-Studio/.agents/skills/calibration-plan/SKILL.md

calibration-plan

Skill4 starsChanged 6 days agoMITView on GitHub

Plans calibration and estimation: parameter sources, priors, moments, targets, identification checks.

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 25
  2. Very specific:

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

    line 27
  3. Distinctive:

    Prefer evidence from comparable economies and periods

    line 47
  4. Distinctive:

    After classification: confirm the split

    line 53
  5. Distinctive:

    After priors: confirm their rationale

    line 54
  6. Distinctive:

    No plan for parameters the data cannot identify

    line 59
  7. Distinctive:

    Every parameter has a classification and a source or prior

    line 67

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

/calibration-plan — Calibration Plan

Category: Scoping & Design · Primary agent: bayesian-dsge-estimation-specialist

Plans calibration and estimation: parameter sources, priors, moments, targets, identification checks.

Arguments: whatever the user wrote after the command name.

Purpose

Plan, before any code is run, how each parameter of a model will be set: calibrated to data, estimated with a stated prior, or fixed at a literature value, with the source and the identification checks that will be used. The output is a plan that prevents ad hoc tuning later and tells reviewers in advance what will count as evidence.

Use it when a model specification is agreed and the team is about to calibrate or estimate it.

Do not use it when the calibration is already done and needs documenting (/calibrate-model) or estimation is in progress (/estimate-dsge).

Before you start

  1. Check required inputs. For each input below record FOUND or ABSENT (never 'assumed'):
    • docs/models/<name>/model-spec.md 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

  • bayesian-dsge-estimation-specialist — primary agent; coordinates the work and integrates results
  • emde-lic-dsge-specialist — Adapts structural models to emerging, developing and low-income economies
  • var-vecm-specialist — Builds VAR, SVAR and VECM models

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
Splitwhich parameters to calibrate, estimate or fixData requirements and identification
Prior familiesbeta, gamma, inverse gamma, normal; endogenous priorsWhat data must override
Targetssteady-state ratios, second moments, IRF matchingDiscipline on dynamics
Identification checkslocal identification, prior–posterior overlap, simulation recoveryCredibility

Workflow

  1. Classify parameters: calibrate, estimate, or fix at literature values
  2. Source each calibrated value (data moment, micro study, published estimate)
    • Prefer evidence from comparable economies and periods
  3. Specify priors with rationale; plan prior predictive checks
  4. List calibration targets and the identification diagnostics to run
    • Include a simulation–recovery exercise for estimated parameters

Checkpoints with the user

  • After classification: confirm the split
  • After priors: confirm their rationale

Common pitfalls

  • Priors chosen to deliver expected results
  • Literature values from different contexts used without comment
  • No plan for parameters the data cannot identify

Output

docs/models/<name>/calibration-plan.md

The plan lists every parameter with its treatment, source or prior with rationale, targets, and the identification diagnostics to run.

Gate / verdict

  • Every parameter has a classification and a source or prior
  • Identification diagnostics are listed for every estimated parameter

Suggested next steps

  • /estimate-dsge
  • /calibrate-model

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.