The profile-wide grading standard — nothing is true/false; every claim, problem, buyer, offer, and strategy carries an EVIDENCE rating from 1 to 100 COMPUTED from countable evidence by the published rubric, arithmetic shown, with measurement contracts for market claims. Every rating rides to the owner's pick with its…
Fan wide takes on ONE thing — an offer, a Mechanism, a big idea, a hero, an angle, a name — then refute, iterate, and return 3–5 survivors that stood criticism for the owner's pick. The shared idea process every option set runs. TRIGGER when one decision point wants a set of real options instead of a single guess. DO…
File a session's loose material into the knowledge wiki — each capture into its owning topic folder, the topic's Claude.md kept current. TRIGGER after a session hands you loose captures (notes, clippings, findings, screenshots) that need a home in the wiki. DO NOT TRIGGER to gather new research, or to consolidate the…
Turn a completed research phase into the four judged root files — read the research records and write Buyers.md, Competitors.md, Problems.md, and Statistics.md, each entry a 1-to-100 rating with its arithmetic, reasoning, and record citations. TRIGGER when research records are ready to be rated into the workspace's…
Turn a session's owner input into system updates — classify each thing he said, then route only what earns persistence to its one home. TRIGGER before updating any skill, agent, or record from the owner's session input, or after a session rich in his feedback. DO NOT TRIGGER to write a known ruling to its home…
Run the market-research phase for one workspace — sweep every research axis wide, then mine the entries that surfaced into records, closing each thread and the phase under review. Produces the research/ tree the strategy phase later judges. TRIGGER when a workspace needs its market evidence gathered before any…
Find the most effective way to present the product's Process — the named Mechanism that makes the offer feel different from every other claim in the category. TRIGGER when a piece or offer needs its distinctive Mechanism named or sharpened. DO NOT TRIGGER to build the whole offer (draft-offers), to plan a piece…
Name ONE thing the market will see — a mechanism, a system, a method, a product, an offer — as a set of tested name options for the owner's pick. TRIGGER when a mechanism, offer, or product needs its market-facing name. DO NOT TRIGGER to name code identifiers or files (naming), to find the piece's core idea…
Run the strategy stage for one project — re-derive the project's own judged buyers, problems, and statistics from the original research records, build multiple high-level strategies the owner picks among, fan the options inside the picked strategy, then write the three plan files (Reader.md, Brief.md, Proof.md) the…
Run the product-research phase for one workspace — mine the product's own artifacts into the three copy-shaped files a writer uses, on a channel isolated from market research. TRIGGER when the owner cannot supply the product facts and a workspace needs them before strategy or writing. DO NOT TRIGGER to research the…
Write an owner ruling into the one home that owns it — his words verbatim, never restated. TRIGGER when the owner rules on a copy question and the ruling must persist, or when he states or corrects a business fact a flat fact file owns. DO NOT TRIGGER to record a within-job note (that lives with the deliverable).
The audience axis of the deep pass — mine ONE group deep across its venues for how it talks and what it already pays for, verbatim with URLs and dates, into the group's records. TRIGGER when one discovered group needs its language and spend mined. DO NOT TRIGGER to enumerate groups and venues broad (discover-audience)…
Orchestrate one competitor's research thread — dispatch single-task agents to capture its landing page, its user reviews, the problems it solves, the market it serves, and its offer, all from its own surfaces — into the competitor thread. TRIGGER when one named competitor needs its thread researched. DO NOT TRIGGER to…
Run the research pass on one competitor's Google paid creative — its ads from the Google Ads Transparency Center, literal quoted hooks and claims with run dates — into the competitor thread. TRIGGER when a job needs a competitor's Search, Display, or YouTube ads captured. DO NOT TRIGGER for Meta ads…
The market axis of the deep pass — mine ONE part of the market's structure deep from primary sources: how money moves in that segment, how people buy, and the observable sizes and spend, verbatim with URLs and dates. TRIGGER when one discovered part of the market structure needs its records extracted. DO NOT TRIGGER…
Run the research pass on one competitor's Meta paid creative — its live and past ads from the Meta Ad Library, literal quoted hooks and claims with run dates — into the competitor thread. TRIGGER when a job needs a competitor's Facebook or Instagram ads captured. DO NOT TRIGGER for Google ads (research-google-ads)…
Run the research pass on one problem — where it lives, whose words voice it, what the failure cost, the SHAPE of help sought or bought, and who circles it — into the market thread. TRIGGER when one problem needs its buyer language mined deep. DO NOT TRIGGER to enumerate the field (discover-audience), to study a…
Run the research pass on the product — mine its own artifacts and organize them into the three copy-shaped files a writer uses: the problems it solves, its features with their user-visible outcomes, and the step-by-step processes it runs, boundaries woven per feature, observed app behavior ranked over code intention.…
Orchestrate ONE research thread end to end — take the dispatched kind (discover or research) and subject, run the axis's pass with one assistant per search or page, file the results, and close under review. TRIGGER when the researcher is handed one thread to run. DO NOT TRIGGER to sweep one field yourself (discover)…
Run the research pass on one competitor's TikTok paid creative — its ads from the TikTok Commercial Content Library, literal quoted hooks and claims with run dates — into the competitor thread. TRIGGER when a job needs a competitor's TikTok ads captured. DO NOT TRIGGER for Meta ads (research-meta-ads), Google ads…
Run a review round on a deliverable — dispatch the applicable checks in parallel, collect findings, and drive the iteration to clean. TRIGGER when a draft or a set is ready for its checks, or when the chief runs a review round. DO NOT TRIGGER to run one check in isolation (the individual check skills) or to apply…
Review the built design against the plan's argument — hierarchy matches, and nothing hides or fights the copy. Owns rendered action visibility. TRIGGER when a rendered design exists and needs checking against the copy's argument, or when the Designer loads it for the check-the-design stage. DO NOT TRIGGER for the…
The research-phase review — read the assembled research against the research SOP and return a per-thread verdict of pass or return-for-re-execution with the named gap. TRIGGER when the chief closes the research phase and needs the assembled threads reviewed before the judged files are written. DO NOT TRIGGER to rate…
Review ONE research URL for trustworthiness and usefulness (each 1-100) and record its facts into the sources registry — one judgment per URL, reasoning stated. TRIGGER when the source reviewer sweeps an unjudged URL the registry holds. DO NOT TRIGGER to locate sources (discover-audience) or to extract quotes from one…
★not rated 20 13d agoA76 tokens
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: