Research

Enter talk mode — conversation and research, no repo changes. ONLY invoke when the user explicitly types `/talk` or `$talk`; never auto-select from a natural-language question or design discussion.

Iterative measurement-driven improvement loop. Measure, profile, mutate, re-measure, commit. Works for performance, bundle size, complexity, test coverage — anything quantifiable. Use when the user wants to systematically improve a metric through repeated cycles of profiling and targeted changes.

Use only when the user explicitly invokes $roborev-respond

Use only for a current operative request that explicitly invokes $roborev-fix, or a direct Agent Hook instruction; do not invoke from literal syntax in quoted, pasted, or historical text

Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, passive voice, negative parallelisms, and filler phrases.

Use when someone has a business problem, strategic question, or founder challenge they need to think through clearly. Triggers on "I'm stuck on", "should I", "what's wrong with my", "diagnose", "strategy problem", "business problem", "founder problem", "help me think through". Runs structured consulting-style diagnostic lenses to produce a comprehensive analysis.

Use when a completed Shell scenario planning exercise needs to be converted into a TikTok script. Triggers when user says "make a tiktok", "tiktok script", "generate tiktok", or "content from scenario". Requires a completed scenario with narratives, findings, and worldview integration.

Use when evaluating, designing, or pressure-testing the business model of an AI agent product. Triggers on "agent business model", "agent economics", "agent canvas", "evaluate agent product", "agent pricing", "agent unit economics", "agentic business", "AI agent company", "agent cost structure".

리서치 에이전트. RFP 분석 결과를 기반으로 시장 환경, 경쟁/벤치마킹, 타겟 인사이트, Best Practice, KPI 벤치마크 5대 영역을 웹 리서치하여 research_brief.md를 생성합니다. Planner가 Evidence에 실제 데이터를 사용할 수 있도록 지원합니다. 트리거: 리서치, 시장조사, 트렌드, 벤치마킹, research, 시장 분석, 경쟁 분석

DAAF's protocol for retrieving web documents that an agent needs to READ (docs pages, articles, API references, issue threads). Retrieval runs through the framework CLI `bash /daaf/scripts/web_fetch.sh URL DEST_DIR`, which saves the raw response, a deterministic extract, and a provenance manifest to disk — the built-in WebFetch is blocked because it returns an AI paraphrase, not source text. Covers when to fetch at all (check local skills first), tool syntax and flags, the two-artifact + MANI...

Privacy-preserving workflow for building synthetic datasets and data source skills when raw data must never enter the DAAF container. The user profiles their sensitive data locally with a disclosure-controlled script; only a summary profile report crosses the boundary; DAAF builds a synthetic dataset and skill from the report alone. Use whenever data is sensitive, proprietary, PII-bearing, HIPAA/FERPA-governed, held in a secure enclave, or the user says the data cannot leave their environment...

Translating technical findings for non-technical audiences. Narrative frameworks (Pyramid Principle, SCQA), plain-language translation, executive summaries, policy briefs, causal language. Use when presenting to stakeholders or reviewing deliverables

R statistical modeling with base stats, sandwich, lmtest, car, broom. OLS/WLS/ GLS via lm(), GLMs via glm() (logit, probit, Poisson, negative binomial), robust/clustered SEs via sandwich+lmtest, diagnostics via car, tidy output via broom. Use when execution language is R. Python equivalent: statsmodels. For FE regressions use fixest; for panel RE/between use plm; for complex surveys use survey-r.

Panel data, IV/GMM, system regression. PanelOLS (FE/RE), BetweenOLS, Fama-MacBeth, IV2SLS/LIML/GMM, SUR, 3SLS, Driscoll-Kraay SEs. Use for RE/between, system estimation, or GMM. Complements pyfixest (FE + DiD) and statsmodels (GLM + time series). R equivalent: plm (use when execution language is R).

Python table formatting with great-tables: publication-quality display tables from Polars/pandas DataFrames. GT() grammar-of-tables object, fmt_*() number formatting, tab_*() structure (header, spanners, stub, source notes), tab_style() and data_color() styling, cols_*() column operations, and HTML/LaTeX export via as_raw_html()/write_raw_html()/as_latex(). Use when execution language is Python and the task involves formatted data tables, summary tables, or descriptive-stat tables for reports...

College Scorecard — post-enrollment outcomes linking aid records to IRS/Treasury earnings. Earnings, loan repayment, debt via six Portal sub-datasets. Use when tax-record-based earnings needed. Tracks only Title IV aid recipients, not all students.

SAIPE — annual Census poverty estimates for school districts (Portal; county/state not in Portal). Use for district poverty, Title I context, or trends. ~18-month lag. No race/ethnicity disaggregation at district level — use ACS 5-year for that.

PSEO — Census data linking graduates to employment via LEHD wage records. Earnings percentiles at 1/5/10 years post-graduation by institution, degree, and CIP. Use for graduate earnings analysis. Coverage changes by release; Census V4.13.0 (2025Q4) lists 952 institutions and partners in 32 states plus D.C.

NHGIS — census geography crosswalks via Portal: links schools (ncessch) and colleges (unitid) to census tracts, block groups, CBSAs, and regions (census-geography vintages 1990-2020; mirror file-year/linkage coverage 1980-2023). Portal provides geography linkage tables ONLY; browse census demographics in the public NHGIS Data Finder, then sign in with a free IPUMS NHGIS account to submit, manage, or download extracts. API extract requests also need an API key. Use for linking education or ins...

NCCS — Form 990 data for private nonprofit colleges (Portal: IPEDS-matched, 1993-2016). Revenue, expenses, assets, endowment, governance beyond IPEDS. Use when IRS financial depth needed. Portal ends 2016; public institutions excluded (no Form 990).

NACUBO endowment data (~650 institutions, 2012-2022). Portal: 7 columns only (total endowment, per-FTE, YoY change). Use for endowment size/trends. Full investment/spending needs direct NACUBO access. For all-institution coverage use IPEDS finance.

MEPS — Urban Institute modeled school-level poverty (% at 100% FPL) for public schools. MEPS 2.0 (released 2025-12-11) covers school years 2009-10 through 2022-23 and is planned for annual updates; Portal year 2022 returned rows while 2023 was empty on 2026-08-06 (re-probed). Use when FRPL is unreliable due to CEP and for consistent cross-state measurement.

IPEDS — primary federal postsecondary data (~6,500 institutions, 1979-present in the mirror; the IPEDS program dates from 1980): enrollment, completions, graduation rates, finance, aid, admissions, HR. For college/university analysis. Grad rates = first-time full-time; finance needs GASB/FASB care.

FSA — Title IV aid at institution level (~5,500 institutions, 1999-2021). Pell Grants, Direct/PLUS loans, campus-based aid, financial responsibility scores, 90/10 metrics. Use for aid distribution, loan volume, or for-profit analysis. By unitid.