SkillAgentSearch skills...

econ-audit

Audit economic analysis outputs (fiscal briefings, macro briefings, market research, longlists, and other quantitative economic documents) against methodology standards, academic literature, and common errors.

Install / Use

npx skills add brycewang-stanford/Auto-Empirical-Research-Skills --skill econ-audit

Installs into whichever agent you are using.

About this skill
📄

SKILL.md

Installable skill definition

Quality Score

95/100

Supported Platforms

Universal

Tags

Our assessment of econ-audit

econ-audit scores 95/100 on our quality scale, 123rd of 1,155 Content & Media skills we index (top 11%).

Its SKILL.md is 19 KB long, well organised into 31 sections with 8 code examples: a thorough specification that gives an agent plenty to work with.

With 4,360 GitHub stars, it is one of the more widely adopted skills in the catalogue.

Substance
30/30
Structure
20/20
Description
15/15
Adoption
15/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated about 2 months ago, so econ-audit is actively maintained.
  • Our last check on 2026-11-15 found the source still online.
  • No license is declared. By default that means all rights are reserved: you can read it, but reusing or redistributing it is not clearly permitted. Ask the author before building on it commercially.
  • Its trust signals score 88/100, with 1 caution from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.

Safety scan

No issues found

Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.

AI review by kimi-k2.7-code on 2026-09-29. Automated pattern scan on 2026-09-28. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.

econ-audit compared with similar skills

All 4 of these similar skills score higher than econ-audit; compare them before choosing.

SkillScoreStarsUpdatedFormat
econ-audit (this skill)by brycewang-stanford954.4k54d agoSKILL.md
siyuanby siyuan-note10047.3ktodayMCP Server
x402by browser-use100121.0k3d agoSKILL.md
algorithmic-artby anthropics100177.9k55d agoSKILL.md
pptxby anthropics100177.9k55d agoSKILL.md

Frequently asked questions

How do I install econ-audit?
Run npx skills add brycewang-stanford/Auto-Empirical-Research-Skills --skill econ-audit. The install tabs above show the steps for each supported agent.
Which AI agents does econ-audit work with?
It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
Is econ-audit safe to use?
Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It declares no license and scores 88/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is econ-audit still maintained?
The repository was last updated about 2 months ago, so econ-audit is actively maintained.

name: econ-audit description: Audit economic analysis outputs (fiscal briefings, macro briefings, market research, longlists, and other quantitative economic documents) against methodology standards, academic literature, and common errors. Runs structured checks across core categories including counterfactual, additionality, discounting, double counting, distributional analysis, Aqua Book RIGOUR, and Flyvbjerg-style strategic misrepresentation detection. Returns a RAG scorecard with issues ranked by severity. allowed-tools:

  • Bash
  • Read
  • Write
  • Glob
  • Grep
  • AskUserQuestion
  • Skill

Only stop to ask the user when: the document type is unrecognised, or fix confirmation is needed under --fix mode. Never stop to ask about: which checks to run (all applicable ones), grading methodology, or output filename.

<!-- preamble: update check -->

Before starting, run this silently. If it outputs UPDATE_AVAILABLE, tell the user: "A new version of econstack is available. Run cd ~/.claude/skills/econstack && git pull to update." Then continue with the skill normally.

~/.claude/skills/econstack/bin/econstack-update-check 2>/dev/null || true
<!-- preamble: project learnings -->

After the update check, run this silently to load prior learnings for this project:

eval "$(~/.claude/skills/econstack/bin/econstack-slug)"
~/.claude/skills/econstack/bin/econstack-learnings-read --limit 3 2>/dev/null || true

If learnings are found, apply them. When a prior learning influences a decision, note: "Prior learning applied: [key]".

Capturing new learnings: After completing this skill, log new insights via:

~/.claude/skills/econstack/bin/econstack-learnings-log '{"skill":"...","type":"...","key":"...","insight":"...","confidence":N,"source":"observed|user-stated|inferred"}'

Types: framework (preferred appraisal framework), parameter (custom overrides), data-source (preferred data), output (past report references), operational (tool/env quirks), preference (formatting/style). Confidence: 9-10 observed/stated, 6-8 strong inference, 4-5 weak. User-stated never decays; observed/inferred lose 1 point per 30 days. All data stored locally. Nothing transmitted.

<!-- preamble: parameter database check -->

After the update check, verify the parameter database is available and check staleness:

PARAMS_DIR="$HOME/econstack-data/parameters"
if [ -d "$PARAMS_DIR" ]; then
  PARAM_COUNT=$(find "$PARAMS_DIR" -name "*.json" 2>/dev/null | wc -l | tr -d ' ')
  echo "PARAMS: $PARAM_COUNT files loaded from $PARAMS_DIR"

  # Check for stale files (last_verified > 2 years ago)
  STALE=$(find "$PARAMS_DIR" -name "*.json" -mtime +730 2>/dev/null | wc -l | tr -d ' ')
  if [ "$STALE" -gt 0 ]; then
    echo "PARAMS_WARNING: $STALE file(s) not updated in 2+ years. Run: cd ~/econstack-data && git pull"
  fi
else
  echo "PARAMS: not found. Using built-in defaults. For full parameter support: git clone https://github.com/charlescoverdale/econstack-data.git ~/econstack-data"
fi

If PARAMS_WARNING appears, tell the user which parameter files may be stale and recommend updating. Continue with the skill normally using whatever parameters are available.

<!-- preamble: safety hooks -->

Safety rules for this skill:

  1. Parameter database is read-only. Never write to, modify, or delete files in ~/econstack-data/parameters/. These are shared, versioned parameters maintained separately. If a parameter needs updating, tell the user to update the econstack-data repo.

  2. Confirm before overwriting. Before writing an output file, check if a file with the same name already exists. If it does, ask the user: "A file named [filename] already exists. Overwrite it, or save with a new name?" Do not silently overwrite.

<!-- preamble: completion status -->

At the end of every skill run, report one of these statuses:

  • DONE: Analysis complete, output generated, all sections finished.
  • DONE_WITH_CONCERNS: Output generated but with caveats (e.g., data gaps, assumptions that need review, sections below expected depth).
  • BLOCKED: Cannot proceed (e.g., missing critical input, parameter database unavailable, framework not supported).
  • NEEDS_CONTEXT: Need more information from the user before continuing.

Format: STATUS: [status] | [one-line reason]

/econ-audit: Economic Analysis Audit

Reviews any economic analysis output against methodology standards and practitioner best practice. Works on fiscal briefings, macro briefings, market research reports, longlists, or any document making economic claims with numbers.

Think of it as a senior partner and an economics professor going through your work and poking holes in it. Returns a RAG (red / amber / green) scorecard with issues ranked by severity, plus an optional --fix mode that proposes concrete corrections.

Arguments

/econ-audit [file or directory] [options]

Examples:

/econ-audit longlist-schools-2026-04-10.md
/econ-audit fiscal-uk-2026-04-10.md --strict
/econ-audit . --fix
/econ-audit macro-uk-2026-04-10.md --framework uk-gb

Options:

  • --strict : Apply higher bars. Green requires all checks pass cleanly; amber becomes red if material.
  • --fix : After scoring, propose concrete corrections for each issue. Optionally apply them (prompted per-issue).
  • --framework <name> : Override the auto-detected framework (uk-gb, eu-brg, wb, adb, au-vic).
  • --section <name> : Emit only one sub-component. Options: full (default), headline (letter grade and count only), issues (issue list only), methodology, fix-plan (if --fix was used). Combinable.
  • --format <type> : Output format(s). markdown (default, always generated), xlsx, word, or pdf.
  • --client "Name" : Add "Prepared for" metadata.

Check categories

The audit runs structured checks across these categories. Checks are framework-aware: some checks apply only to specific frameworks.

| Category | Checks | Applies to | |---------|--------|-----------| | Counterfactual | Is Do Nothing specified? Is it dynamic (not static)? Are all benefits incremental to it? | All | | Additionality | Deadweight, displacement, leakage applied? Rates defensible? Not applied to carbon? | All | | Discounting | Framework-appropriate rate? Declining vs flat correct? Health at 1.5% if applicable? | All | | Optimism bias | Applied at the right rate for project type and stage? | uk-gb only | | Double counting | Construction jobs + capex? Journey time + land value? Gross earnings + tax? | All | | Transfers | Tax receipts, welfare payments correctly excluded from social CBA? | All | | Sunk costs | Prior spend excluded? | All | | Cash vs non-cash | Economic NPV includes all; Financial NPV includes only cash in / cash out? | All | | Five Case consistency | Financial vs Economic capex, benefits register vs realisation, risk register vs contingency, preferred option consistent | uk-gb, au-vic business cases | | Distributional | Is distributional incidence reported? Is it explicit for uk-gb if material, and mandatory for wb, adb, au-vic? | Framework-dependent | | Aqua Book RIGOUR | Repeatable, Independent, Grounded, Objective, Uncertainty-managed, Robust | uk-gb | | Flyvbjerg indicators | Scope near thresholds, low contingency, optimistic demand, benefit / cost asymmetry | All | | Data vintages | All cited numbers have a vintage within a reasonable window | All | | Source quality | Sources cited, from official or peer-reviewed origins | All |

Instructions

Step 1: Identify the document

If the user passes a single file path, load that file. If they pass a directory, glob for markdown files and audit each one in turn.

Auto-detect the document type from the file name or the KEY NUMBERS block at the top:

  • longlist-*.md, type: longlist → Longlist
  • macro-*.md, type: macro-briefing → Macro briefing
  • fiscal-*.md, type: fiscal-briefing → Fiscal briefing
  • market-*.md, type: market-research → Market research
  • briefing-*.md, type: briefing-note → Briefing note
  • Unknown → ask the user what it is once.

Auto-detect the framework from the document header or KEY NUMBERS. Default to uk-gb if missing.

Step 2: Run all applicable checks (silent)

For each check in the category table above, evaluate:

  • Pass (green): the check is met or clearly met.
  • Warn (amber): the check is partially met or cannot be determined from the document.
  • Fail (red): the check is clearly violated or entirely absent when it should be present.

Checks are framework-aware. Do not apply optimism bias checks to a non-uk-gb document. Do not apply Aqua Book RIGOUR to a non-uk-gb document. Do apply Five Case consistency to uk-gb and au-vic business cases, and skip for other document types.

Step 3: Score and rank

Compute:

  • Count of passes, warnings, fails
  • Overall letter grade: A (no fails, few warnings), B (few fails, some warnings), C (several fails, many warnings), D (serious methodology gaps), F (unusable)
  • Under --strict: amber escalates to red for material issues.

Rank issues by severity:

  • Critical: methodology gap that invalidates the conclusion (missing counterfactual, wrong discount rate, material double counting, transfers included in social NPV).
  • High: material gap that weakens the conclusion but does not invalidate it.
  • Medium: presentation or evidence gap that should be addressed before submission.
  • Low: stylistic or completeness nit.

Step 4: Fix plan (if --fix)

For each critical and high issue, propose a concrete fix. Example fixes:

  • "Add a dynamic counterfactual describing deterioration of existing assets over the appraisal period."
  • "Apply 24% optimism bias uplift to capital costs per HMT Table 1 for this project type and stage."
  • "Remove council tax receipts from the benefits table; note in a separate fiscal annex."
  • "Compute additionality using 20% deadweight, 25% displacement, 10% leakage; current analysis uses unadjusted gross benefits."

Under --fix, for each proposed fix, ask the user to confirm application. If confirmed, edit the source document in place. If not, record in the fix plan and move on.

Step 5: Write the output

Save audit-[target-slug]-[YYYY-MM-DD].md with this structure.

<!-- KEY NUMBERS
type: econ-audit
target: [target file or directory]
framework: [framework]
strict: [true|false]
passes: [count]
warnings: [count]
fails: [count]
grade: [A|B|C|D|F]
critical_issues: [count]
high_issues: [count]
fixes_applied: [count]
date: [YYYY-MM-DD]
-->

# Audit: [target file name]

**Date**: [YYYY-MM-DD] · **Framework**: [framework] · **Strict mode**: [on/off]

## Headline

**Grade**: **[A / B / C / D / F]**
**Checks**: [passes] pass, [warnings] warn, [fails] fail
**Critical issues**: [count]

[One-sentence summary: e.g. "Grade B. The CBA is sound in structure but applies additionality inconsistently and is missing a distributional analysis."]

## Scorecard

| Category | Result | Notes |
|----------|:------:|-------|
| Counterfactual | [G/A/R] | [one line] |
| Additionality | [G/A/R] | [one line] |
| Discounting | [G/A/R] | [one line] |
| Optimism bias (if uk-gb) | [G/A/R] | [one line] |
| Double counting | [G/A/R] | [one line] |
| Transfers | [G/A/R] | [one line] |
| Sunk costs | [G/A/R] | [one line] |
| Cash vs non-cash | [G/A/R] | [one line] |
| Five Case consistency (if applicable) | [G/A/R] | [one line] |
| Distributional | [G/A/R] | [one line] |
| Aqua Book RIGOUR (if uk-gb) | [G/A/R] | [one line] |
| Flyvbjerg indicators | [G/A/R] | [one line] |
| Data vintages | [G/A/R] | [one line] |
| Source quality | [G/A/R] | [one line] |

## Issues (ranked by severity)

### Critical

- **[Issue 1]**: [description] · Location: [section / line] · Fix: [suggested fix]
- **[Issue 2]**: [description] · Location: [section / line] · Fi

Truncated for display — read the full file on GitHub.

Related Skills

View on GitHub
GitHub Stars4.4k
CategoryContent
Updated1mo ago
Forks527

Languages

Stata

Trust signals

88/100

From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.

1 medium
econ-audit — Universal Skill: Install & Safety Check | SkillAgent