research-competitors

Competitor research tells the marketer what to do next; it is not a list of ads. Craft: references/research-spec.md; worker briefs: its step skills (research-competitors-<step>); the report: references/report-format.md — next to this file.

The rule that matters most: free pass first, ask nothing, report once (../marketing-guide/research-pattern.md (in the marketing-guide skill) — read it before the first search). For a full teardown, research every dimension for every main rival, browser-only sources included, before the first message; never ask permission for free work or offer to pay for what a free read could get. An ask that names an area ("what ads are they running?") gets that area covered completely, news and launches first.

Before you start

1. Setup first: do the marketing-guide skill (../marketing-guide/SKILL.md) §1–2 (connect, brand, brain read) before anything else; the job line is "I research your competitors and save what I find through it." Here: a named brand not in the account → create it from its website (gooseworks:brand_setup_create) without asking and wait for its research (gooseworks:job_get); don't show a menu unrelated to research. Fallback brand read: gooseworks:brand_read (summary, kit, products, competitors, organic, learnings). Either path: gooseworks:competitor_read (+ suggestions), gooseworks:automation_list; workspace: research/competitors/ (last report, rival records), research/market/ (the newest market report and its "which market" decision), research/audience/ (buyer-groups.md and the latest audience report), plans/, reports/performance/. Brand facts: brain first, website second. Claims: only the read step of the claims-gate skill (../claims-gate/SKILL.md) now.
2. Focused output: for a battlecard, page tools/pixels, or LinkedIn engagers, select and finish the matching branch in references/focused-outputs.md after setup, then return. Reuse saved evidence; these branches replace the full teardown below.
3. Run memory: job competitor research or competitor watch; what this job hands the atom is in references/run-memory.md. Its "since last time" opens the chat and the report's story.
4. Never change the host's settings or permissions, and never seed a search or a worker with names or claims from memory — only what a source gave you.

Steps

1. Find the field yourself, before any worker starts. About 6–10 searches by the main run: news for our brand and the category (last 90 days), then "best <category>", "<brand> alternatives", "<the strongest saved rival> alternatives", "<category> tools", and one AI answer. The saved roster is a starting list, never the answer. State the frame of reference in one line — the category we compete in (research-market's decision when it exists) and the other frames we could claim (e.g. "AI ad-creative tool" vs "AI agents for go-to-market teams"), because the frame decides which rivals matter. Then choose 3–6 main rivals — the ones buyers compare us with now, each with its reason — and 2–4 adjacent options (doing it yourself, an agency, the platforms' own tools). If the frame or the main rivals differ from the last run's, say so in one line with the reason ("Last run compared us with Jasper and Copy.ai; this run with Creatify, because…") — the frame never swings silently. A name you'd only watch goes in the brief competition line, not as a main rival. Anything from the last 30 days is a candidate timely opening.

2. Brief the workers — lean. Exactly one subagent per main rival with the research-competitors-rival-pass skill (../research-competitors-rival-pass/SKILL.md) (a new entrant you'd call closest to our buyer, or hitting us harder than a main rival, counts as a main rival — it gets its own worker, never just a name), one with the research-competitors-category-pass skill (../research-competitors-category-pass/SKILL.md), one with rival-pass on our own brand — no more, no workflow on top. Paste the short brief's full text with brand_id, our product summary, approved claims, the frame, a typical-buyer scenario for pricing ("a 10-person team making 40 ads a month"), the off-limits list and today's date; workers don't load skills or start their own subagents. Standard depth is the brief's cap (about 12 searches and 12 fetches each); a deep teardown — quarterly, or when the marketer asks — doubles it. A cheaper model is fine. Workers use web search, page fetch and the free server reads, and return everything, including the pages that need the browser. Edge cases — no browser, a stalled or stopped worker, facts from earlier research, a bad rival on the list, a report already saved today, a late read — follow the research-competitors-edge-cases skill (../research-competitors-edge-cases/SKILL.md).

3. Open the browser-only sources yourself, before the report. Workers often can't. For every main rival and for us: the Meta, LinkedIn, Google and TikTok ad libraries, searched by the advertiser's page or name (not the domain), plus the pages workers listed (X and LinkedIn posts, script-only pricing pages, review pages that block fetch). Tell the marketer once, in one line, first.

4. Go deep on what moved. For the 1–2 biggest launches, watch the demo and the 3 best third-party posts (references/rival-craft.md §5) and break each into hook, setup, reveal, proof, CTA; write a format brief per demo. Counter-position only if the leverage test passes — we are the challenger here, we genuinely do the same job better, we have dated proof, and our answer can reach enough of the rival's buyers to matter; then draft the counter-position kit (references/angles-and-gaps.md "Launches"). Otherwise it goes under "What we're not doing": not every launch needs an answer. For each timely opening, draft the move (spec §4) with an end date.

5. Write the story in the shape of references/report-format.md, formats and angles from references/angles-and-gaps.md. A new entrant that surfaces only now (from the category pass) and is closest to our buyer gets a short rival pass before the report (the research-competitors-rival-pass skill (../research-competitors-rival-pass/SKILL.md) items 1, 4, 5, 6, 7) — never a profile-less "closest rival". Check every number, date, quote and "no ads found" against what a worker or the browser returned; drop what you can't trace.

6. Check, then deliver. Run the check in references/report-format.md, then deliver through the write-report skill (../write-report/SKILL.md) (topic competitor-research; brand copy research/competitors/<YYYY-MM-DD>-report.md, keep-history). Also save, in the shapes in references/report-format.md, each rival record, landscape.md and angle-bank.md (supersede) through the save-to-brain skill (../save-to-brain/SKILL.md), and swipe items to the inspiration library (gooseworks:social_inspiration_save).

7. Save the run — before the closing message. The End steps of the run-memory skill (../run-memory/SKILL.md), with the numbers and recommendations in references/run-memory.md. When the marketer answers a decision afterwards, re-save the same day's record with it (references/run-memory.md).

8. One message: the draft and the decisions, in the closing-message shape in references/report-format.md, with the one-line "saved this run's notes" folded in (at most five one-line decisions): the one thing to do first and the link; which openings to act on · the rivals covered (add or drop any?) · sign-off for the claims the top moves need · a paid top-up only for what free reads and the browser couldn't get, named, with the estimate and one yes to "up to N credits" (the spend-gate skill (../spend-gate/SKILL.md) branch B) · a repeat, only if it would help this brand (marketing-guide §3 Schedules; a quiet field: quarterly, or none), saying that scheduled runs have no browser, so ad libraries are read only in runs you start (references/watch-and-react.md). Optional, one line each: tracking new rivals (gooseworks:competitor_save), keeping the top angles as rules (the save-learning skill, ../save-learning/SKILL.md).

9. Watch and react (references/watch-and-react.md, same dimensions). What changed (on ask, or the scheduled repeat): opens with what changed against the stored numbers and each rival's snapshot; a report only when something crosses a threshold, a run record every time. React now ("X just launched / shut down"): verify, read the market's reaction, alert this turn with the window, the move and a draft brief.

Approval gates

• The free pass needs no yes and asks nothing; a paid top-up needs one yes to a ceiling; schedules are created paused.
• Only with the marketer's yes: newsletter or trial sign-ups, tracking a rival, saved rules, new claims.
• Research delivers reports, battlecards and briefs; makers produce ads and posts. Never publish, contact anyone, copy a rival's words or footage, mock a rival or its customers, or claim their spend or results. Named comparisons need sign-off and dated proof.

Output and write-back

Paths and readers: the research-competitors rows of the marketing-guide's §4 table. Each file's title, tags and save mode: references/report-format.md; what-changed reports: references/watch-and-react.md; focused outputs: references/focused-outputs.md.

Done when

Focused outputs use their branch's completion check (references/focused-outputs.md). For a full teardown:

• No question before the report, beyond picking a brand.
• The frame of reference is stated; rivals were chosen from category evidence; a change of frame or rivals since the last run is stated with its reason.
• News covered every rival and our brand.
• Every ad library was searched for every main rival and us — in the browser where workers couldn't, or the report says once which couldn't be opened and why.
• Every rival has a same-shape profile, with a praise and a complaint — including any rival the report calls closest to our buyer.
• The report passes write-report §3 and the report-format check; its first move is not table stakes; it has the baselines table, leads with timely openings, counter-positions only where the leverage test passed, states disagreements with sibling reports, and ends with what it means for positioning.
• It was saved to the brand and opens locally.
• The run record was saved before the closing message, with this run's moves as recommendations and the last run's outcomes.
• The closing message held only decisions, and a repeat was offered only with a reason it would help.

Out of scope

Making the ads (make-static-ads; video ads: the goose-video skill) · budgets and channel bets (plan-campaign) · what people say about our brand and the market (research-market) · our own search visibility (improve-search-visibility) · finished positioning and messages (research-market).