MVPavan/via / .claude/skills/research
research
VerifiedUse when a decision needs multiple sources gathered and weighed. Answer a single factual lookup directly from an appropriate source.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 2.1 KB |
| agents/openai.yaml | config | 42 B |
| references/examples.md | markdown | 2.7 KB |
| references/frameworks.md | markdown | 3.1 KB |
| references/playbooks.md | markdown | 3.8 KB |
| references/source-selection.md | markdown | 2.2 KB |
| references/synthesis-engine.md | markdown | 1.9 KB |
references/examples.md
View on GitHub# Examples
Use these as quality calibration only.
## Tech Tooling
Weak:
"Temporal is good for workflows, Celery is good for tasks, and Redis queues are simple."
Strong:
"Temporal wins when workflow correctness depends on durable timers, signals, retries, and replay across worker crashes. That strength is also its cost: it adds a workflow service, deterministic workflow constraints, and a new operational model. For a single-user assistant with human approvals, Postgres-backed jobs may be enough until runs span days or require compensation semantics. The upgrade trigger is not 'we have workflows'; it is 'we cannot reliably resume or audit long-running workflows with our current job model.'"
## Investing
Weak:
"AI demand is increasing, so semiconductor companies may benefit."
Strong:
"The investable question is not whether AI demand grows; it is where incremental dollars land and when that shows up in fundamentals. GPU vendors capture direct capex first, but power, networking, memory, cooling, and data-center REITs may see delayed second-order effects. The thesis needs segment exposure, margin sensitivity, backlog quality, customer concentration, and whether the growth is already priced in."
## Product
Weak:
"Users want collaboration features, and competitors have them."
Strong:
"The collaboration request is actually two jobs: shared review before publishing and async accountability after publishing. Competitors bundle both into generic comments, which creates noise. A better product decision is to separate approval workflow from discussion: approvals need state, owners, and audit trail; discussion needs context and low friction."
## Developer Tools
Weak:
"Use the SDK because it has tools and handoffs."
Strong:
"The SDK should own the loop only if its abstractions match your failure model. If you need provider portability through LiteLLM, custom approval gates, and first-class artifact logging, keep the orchestration boundary yours and adapt SDK utilities selectively. The SDK is a component, not the architecture."
## Policy
Weak:
"The law requires compliance, and penalties may apply."
Strong:
"Separate three layers: written rule, enforcement pattern, and practical compliance norm. If enforcement actions target deceptive marketing rather than technical paperwork defects, the highest-risk control may be claims review, not form automation. That distinction changes the product requirement."
## Quick Tests
- Replacement test: if the paragraph could apply to any industry, make it specific.
- Decision test: if it does not help a decision, cut it.
- Source test: if challenged, can the report show where the claim came from?
- Scope test: if it introduces a new research direction, did the user approve it?
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| 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: 115k | Updated | Claude CodeCodex | Verified | Install |
| add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| 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: 115k | Updated | Claude CodeCodex | Verified | Install |
| 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: 115k | Updated | Claude CodeCodex | Verified | Install |