radarist.ai
Official@radarist · Spain
Radarist is an open-source, local-first innovation platform that turns weak signals into decisions. It runs on your machine, and the knowledge it builds is your
Agent Skills by radarist.ai
Showing 69 vetted skills indexed across 1 GitHub repositories.
radaristexpected-value-decision-tree
Rolls back decision trees to expected values and prices information with EVPI, EVSI, and tornado sensitivity.
radaristfive-forces-analysis
Assesses industry structural profitability using Porter's Five Forces with evidence-rated indicators.
radaristoss-project-health
Assess open-source repository maintenance vitality using CHAOSS metrics and OpenSSF Scorecard checks.
radaristreference-class-forecasting
Sanity-checks inside-view estimates against outcome distributions of comparable past projects.
radaristvalue-intellectual-property
Values early-stage patents, software rights, and know-how using cost, market, income, and rNPV approaches.
radaristthree-horizons
Classifies portfolio bets into H1, H2, and H3 growth horizons with horizon-appropriate evidence bars.
radaristdecision-matrix-mcda
Ranks options against weighted criteria using MCDA with AHP weights and sensitivity analysis.
radaristkey-assumptions-check
Surfaces and challenges the stated and unstated premises behind a conclusion.
radaristbenchmark-model-claims
Audits model benchmark claims across six integrity domains and emits a 0-5 reliability score.
radaristabstain-or-escalate
Decides how to handle claims that verification could not confirm.
radaristjtbd-framing
Frames vendor and technology comparisons around customer Jobs to be Done using Ulwick's outcome grammar.
radaristgrounded-fact-check
Verifies load-bearing claims in finished drafts against live sources and emits a correction ledger.
radaristamstar2-review-appraisal
Appraises systematic reviews with the 16-item AMSTAR 2 checklist and derives overall confidence ratings.
radaristevidence-appraisal
Rates certainty of a body of evidence per outcome using the GRADE methodology.
radaristmeta-analysis
Pools effect sizes from multiple studies into fixed-effect and random-effects estimates with heterogeneity statistics.
radaristskill-name-here
Provides a reusable template for authoring structured analytic method skills.
radaristanalyze-patent-claims
Parses one patent's claim set into dependency trees, transitional-phrase scope, and drafting flags.
radaristexperimental-design
Produces preregisterable experiment design cards with power analysis and validity audits.
radaristindicators-validation
Validates scenario indicator lists by rating diagnosticity and building monitoring plans.
radaristbackcasting
Builds a dated milestone chain backwards from a desired end state to the present.
radaristposition-competitor
Maps competitors onto a two-axis positioning landscape with orthogonality tests and whitespace analysis.
radaristestimative-language
Rewrites and lints likelihood and confidence wording against ICD 203, PHIA, or IPCC standards.
radaristevolution-stage
Tags technologies with Wardley evolution stages and evidence-based ways of working.
radaristrate-source-admiralty
Grades sources on the NATO Admiralty Code reliability and credibility axes.
Frequently Asked Questions About radarist.ai
FAQPage SchemaWhat tasks can I accomplish with Radarist's skill library?▼
You can run structured decision and research methods: decision trees with EVPI/EVSI, Porter's Five Forces, Wardley mapping, scenario planning, Bayesian updating, meta-analysis, GRADE evidence appraisal, patent claim and landscape analysis, market sizing (TAM/SAM/SOM), benchmark-claim audits, and pre-publication fact-checking with citation validation.
Who is Radarist built for?▼
Radarist targets analysts, strategists, researchers, and investors who need defensible reasoning: intelligence-style tradecraft (ACH, key assumptions checks, Admiralty source grading), foresight practitioners (horizon scanning, Delphi, backcasting), and R&D or venture teams evaluating technology readiness, IP value, and market attractiveness.
Is Radarist free to use and under what license?▼
Yes. Radarist is open-source under the MIT license for skill text. Some embedded instruments carry their own terms noted per skill, such as AMSTAR 2 (CC BY 4.0), Cochrane RoB 2 (CC BY-NC-ND 4.0), the UK PHIA yardstick (OGL v3.0), and Ecosyste.ms data (CC BY-SA 4.0).
How does Radarist run and where does my data go?▼
Radarist is local-first: it runs on your own machine and the knowledge it builds stays with you, per its platform description. Skills are invoked through natural-language trigger phrases such as "red team this claim", "score this on TRL", or "build a weighted decision matrix".
How do the skills handle claims that cannot be verified?▼
Verification skills follow explicit abstention discipline: grounded-answer ships only source-checked claims, grounded-fact-check returns a confirmed/corrected/unverifiable ledger, and abstain-or-escalate decides whether to drop, refuse, escalate, or report conflicting sources rather than shipping a hedged guess.