Skip to content

davepoon/buildwithclaude / plugins/all-skills/skills/competitor-research

competitor-research

Verified

Research competitor products, launch history, and acquisition evidence from public sources or supplied materials. Use when evaluating growth claims, comparing competitors, or choosing growth experiments. Produces a cited brief that separates observations, attributed claims, estimates, and hypotheses.

FileKindSize
LICENSEother1.1 KB
SKILL.mdskill md6.7 KB
references/evidence.mdmarkdown3.1 KB
references/report-template.mdmarkdown1.0 KB

references/evidence.md

View on GitHub
# Evidence and growth interpretation

## Claim ledger

Keep a lightweight ledger for important claims:

| Claim | Label | Source | Event date | Published / observed | Confidence and caveat |
|---|---|---|---|---|---|
| Public launch occurred | Observed announcement | URL or filename | Date stated in source | Publication / retrieval date | Announcement proves an announcement, not first availability |

Primary sources are strongest for what a company currently offers. A founder's explanation is an attributed account of causality, not an independent experiment. Provider traffic estimates can establish a hypothesis, not actual visits. A third-party listing may establish discoverability, not submission date, customers, or ownership.

## Common traps

- **Historical timing:** a directory migration or backlink crawler refresh can give many links the same first-seen date. Check actual page dates or archives before calling it a coordinated launch.
- **Missing early evidence:** earliest available mention is a lower bound on the observed history, not proof of when the product began. Domain age is especially weak if ownership changed.
- **Revenue definitions:** record currency, measurement period, gross/net, recurring/one-off, source, and product scope when available. If a dashboard mixes lifetime sales and MRR, keep them separate. A payment-provider badge alone does not establish channel attribution.
- **Distribution metrics:** list directories as submitted, accepted, or live, based on evidence. Keep package downloads, skill installation telemetry, unique accounts, active use, and paying users separate.
- **Acquisition attribution:** an installation count plus MRR does not yield a conversion rate without matched cohorts, dates, and user identities. Growth after a launch suggests a test, not proof of the mechanism.
- **Costs:** estimate the user's implementation and ongoing data/support costs separately. Free instructions can still require a paid host/model; free API reads may still require authorization.
- **Comparisons:** do not compare launch-month revenue with current revenue without labeling dates. Distinguish company totals from the product being studied.

## Evidence sufficiency

One reliable source may establish a pricing or feature fact. A growth-causality claim needs stronger attribution evidence; otherwise present it as a hypothesis even if many articles repeat it.

When the site cannot be reached, explain whether you have independent evidence or only supplied material. Do not claim to have visited an inaccessible page. When there are too few facts for a timeline, replace it with a short evidence inventory and useful next questions.

## Example of a bounded inference

“The project publishes a no-account skill and an authenticated paid data connector” can be established from current product documentation.

“The free skill drove most paid subscriptions” requires attribution that documentation and installation counts do not provide. A reasonable experiment is to offer a useful free workflow and measure downstream activation, rather than copy an assumed conversion rate.
add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks.Stars: 115kUpdated Claude CodeCodexVerifiedInstall