research_skills

In the author's own words, from each skill's SKILL.md.

• audit-review

  Mechanical audit-and-fix pass over a completed `review` run. Checks (a) completeness — every stage ran, and every printed equation, table, figure, headline claim and deposited script was covered — and (b) mechanical errors — scripts re-run clean, paper numbers match print, reproduced numbers come from the right row, anomalies reach the claims they bear on, every number traces to a log. Rates no c…

  audit-review/SKILL.md on GitHub (opens in a new tab)

• review-digest

  Compile a completed `review` run into the evaluator-facing issue ledger (`ISSUES.md`) and render it as a self-contained HTML digest — a ranked, categorized issue list for human evaluators. Use when asked to "make the reviewer digest", "write the issues file", "render the review", or produce an evaluator-facing version of a review.

  review-digest/SKILL.md on GitHub (opens in a new tab)

• review

  Review and replication of an academic paper. Reads the paper and its version history, audits the code against the text, executes the replication package, audits how the data were built, challenges every finding in a fresh context, and records what the evidence establishes claim by claim. Use when asked to review a paper, audit a draft, or start a replication.

  review/SKILL.md on GitHub (opens in a new tab)

• ai-sweep

  Sweep a piece of writing for machine-register prose AND for exposed plumbing — stock AI filler ('it is worth noting', 'delve', evaluative adverbs), rhetorical scaffolding, fact-free sentences, and deliberation leak (self-narrating verification, tours of rejected alternatives, pre-empted objections, process autopsies, stale hedging, trivial-derivation apparatus). Two layers: a mechanical tell pass…

  review-digest/ai-sweep/SKILL.md on GitHub (opens in a new tab)

• challenge

  Fourth review stage, run by a fresh agent: test the review's evidence in both directions — where the review errs against the authors and where it errs in their favour — rebuild it from the pristine deposit and the printed page, and rule on each element: STANDS, STANDS WITH CHANGES, WRONG or ADD, with proposed wording and every adverse result the stage's own scripts produced. Rules on evidence, ne…

  review/challenge/SKILL.md on GitHub (opens in a new tab)

• claims

  Third review stage: work up from the surviving estimates to the narrowest story each can carry, test that story, and build the claim ledger — one row per load-bearing claim recording how central it is, whether it holds, what survives and what is not established, and the strongest defence. The ledger records what is true; it rates nothing.

  review/claims/SKILL.md on GitHub (opens in a new tab)

• code-and-data

  Second review stage: read the code against the text (with the formula map), run the replication package and reproduce its anchors, audit how the load-bearing variables were built from their sources, and stress-test every result family with the fragility battery and its design-specific modules.

  review/code-and-data/SKILL.md on GitHub (opens in a new tab)

• paper

  First review stage, no code: the paper's version history (what moved between drafts and print), the printed-table consistency pass (do the printed cells support the prose?), and the research-design critique (identification, confounders, controls).

  review/paper/SKILL.md on GitHub (opens in a new tab)

• verdict

  Fifth review stage, run by one fresh agent: read every finding and the challenge reports, decide for each claim whether it holds, what survives and what is not established, assemble the design-level findings, and write SYNTHESIS.md. Records what is true; rates nothing.

  review/verdict/SKILL.md on GitHub (opens in a new tab)

• controls-samples-diagnostics

  Diagnostics for controls, samples, and denominators that define the estimand: mediator controls, control horse races, same-source covariates, covariate-by-period under fixed effects, treated shares from the regression sample, and denominator transportability. Invoke from `code-and-data` whenever a horse race, mediator, attribution claim, subset restriction, or same-source covariate touches a targ…

  review/modules/controls-samples-diagnostics/SKILL.md on GitHub (opens in a new tab)

• cross-sectional-diagnostics

  Diagnostics for cross-sectional observational designs: functional form, common support, influence and leave-region-out, dependence structure including geography, negative controls, and omitted-variable-bias robustness values benchmarked against named covariates. Invoke from `code-and-data` for cross-sectional claims resting on a control set.

  review/modules/cross-sectional-diagnostics/SKILL.md on GitHub (opens in a new tab)

• did-event-study-diagnostics

  Diagnostics for difference-in-differences and event-study designs: treatment-timing inventory, estimator validity under heterogeneous effects, pre-period paths, anticipation, alternative comparison groups, and terminal-period sensitivity. Invoke from `code-and-data` whenever a target result compares units across treatment timing.

  review/modules/did-event-study-diagnostics/SKILL.md on GitHub (opens in a new tab)

• experiment-diagnostics

  Diagnostics for randomized experiments: assignment integrity, attrition and missingness, intent-to-treat estimation, prespecified outcome families with multiplicity control, randomization inference, and noncompliance handling. Invoke from `code-and-data` whenever a target result rests on randomized assignment.

  review/modules/experiment-diagnostics/SKILL.md on GitHub (opens in a new tab)

• iv-diagnostics

  Diagnostics for instrumented specifications: first-stage strength at the design's inference level, reduced form, weak-instrument-robust inference, exclusion threats, and omitted-variable-bias sensitivity on the exclusion restriction. Invoke from `code-and-data` whenever any target specification uses an instrument. Strength is not exogeneity.

  review/modules/iv-diagnostics/SKILL.md on GitHub (opens in a new tab)

• panel-diagnostics

  Diagnostics for panel designs without a discrete treatment event: unit and time dependence, serial-correlation-robust inference, fixed-effect structure sensitivity, balance and attrition, and group-level leave-outs. Invoke from `code-and-data` for panel variation that is not an adoption-event design.

  review/modules/panel-diagnostics/SKILL.md on GitHub (opens in a new tab)

• quantitative-model-diagnostics

  Diagnostics for calibrated and quantitative-theory results — a headline number ("X explains 40% of Y") produced by solving or simulating a model. Inventories the free parameters, sweeps the calibration over defensible ranges, tests fit on non-targeted moments, and checks the code solves the model the text describes. Invoke from `code-and-data` whenever a load- bearing magnitude comes out of a cal…

  review/modules/quantitative-model-diagnostics/SKILL.md on GitHub (opens in a new tab)

• rdd-diagnostics

  Diagnostics for regression-discontinuity and regression-kink designs — sharp, fuzzy, discrete-score, multi-cutoff and geographic: which output row carries the verdict, bandwidth sensitivity, density and manipulation tests with manipulation-robust bounds, covariate continuity and covariate adjustment, placebo cutoffs, donut checks, RD plots, local-randomization inference, boundary-discontinuity es…

  review/modules/rdd-diagnostics/SKILL.md on GitHub (opens in a new tab)

• scale-sensitive-diagnostics

  Diagnostics for scale-sensitive transforms and count outcomes: unit rescaling inside ln(1+x) and asinh, count models on the full sample, count versus per-capita specifications, unstable ratios, and placebos in the corrected specification. Invoke from `code-and-data` whenever ln(1+x), asinh, counts, per-capita scaling, or weak denominators carry a target result.

  review/modules/scale-sensitive-diagnostics/SKILL.md on GitHub (opens in a new tab)

• shift-share-diagnostics

  Diagnostics for shift-share and exposure-weighted designs, including designs the paper does not label shift-share: Rotemberg-weight concentration, effective number of shocks, decomposition identity, placebo-shock rejection, and shock-level inference. Invoke from `code-and-data` whenever any regressor or instrument is an exposure-weighted sum of shocks.

  review/modules/shift-share-diagnostics/SKILL.md on GitHub (opens in a new tab)

• small-n-diagnostics

  Diagnostics for small samples and minimum detectable effects: MDE at conventional power against claimed and plausible effect sizes, lean re-tests when saturated controls overfit, and estimator-specific small-sample artifacts. Invoke from `code-and-data` whenever units or clusters are few, or a null carries interpretive weight.

  review/modules/small-n-diagnostics/SKILL.md on GitHub (opens in a new tab)

• synthetic-control-diagnostics

  Diagnostics for synthetic control and synthetic difference-in-differences: pre-period adequacy, placebo-unit rank, weight diagnostics including lagged-outcome degeneracy, terminal-period sensitivity, and in-time placebos. Invoke from `code-and-data` whenever a target result uses a synthetic comparison unit.

  review/modules/synthetic-control-diagnostics/SKILL.md on GitHub (opens in a new tab)

• theory-diagnostics

  Diagnostics for formal-theory results — theorems, propositions, proofs, derivations, closed-form claims. Re-derives from the paper's own primitives, constructs counterexamples, stress-tests the assumptions, and checks that each proof delivers exactly what it claims. Invoke from `code-and-data` whenever a load-bearing result rests on a proof or derivation rather than an estimate. A kill here is a…

  review/modules/theory-diagnostics/SKILL.md on GitHub (opens in a new tab)

• time-series-diagnostics

  Diagnostics for regressions on trending or highly persistent time series: unit-root order (opposite-null), cointegration (residual, system, reversed-null, break-robust), the Granger–Newbold R²-vs-Durbin–Watson fingerprint, Prais–Winsten and first-difference coefficient survival, an order-agnostic ARDL/ECM bounds test, out-of-sample validation, and calibrated size/power. Invoke from `code-and-data…

  review/modules/time-series-diagnostics/SKILL.md on GitHub (opens in a new tab)

• whimsical-assumptions-diagnostics

  Diagnostics for the premises of a model-based conclusion: the assumptions a theorem, a calibrated number, a model-interpreted estimate or a policy prescription stands on. Inventories the load-bearing assumptions (stated, structural, closure, exclusion, bridge), grades each one's empirical warrant from sources, and relaxes each in a realistic direction to see whether the conclusion survives. Invok…

  review/modules/whimsical-assumptions-diagnostics/SKILL.md on GitHub (opens in a new tab)

No skill matches that filter.

Similar skills

• reaby morluto Reverse engineer anything with agents, from app behavior down to native binaries. 412k stars · 5 skills
• superpowersby obra An agentic skills framework & software development methodology that works. 309k stars · 15 skills
• skillsby mattpocock Skills for Real Engineers. Straight from my .agents directory. 307k stars · 30 skills
• ECCby affaan-m The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Cod… 293k stars · 30 skills
• andrej-karpathy-skillsby multica-ai A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls. 227k stars · 1 skill
• skillsby anthropics Public repository for Agent Skills 185k stars · 19 skills