# __SKILL__

> REPLACE ME with 2-4 sentences a router can act on. State what the skill produces, then name the concrete phrasings that should load it — 'screen these abstracts', 'build a PRISMA flow', 'check this corpus for missing seminal papers'. Then name the near misses that must NOT load it. Write triggers, not marketing.

- Skill: `timsmykov/skill` (Agent Skill, multi-file: 5 files)
- Install (CLI): `npx skillmds@latest add timsmykov/skill`
- Raw SKILL.md: https://api.skillmd.com/api/skills/timsmykov/skill/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Marketing & Growth
- Author: timsmykov (https://skillmd.com/u/timsmykov)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/timsmykov/skill

---


# __SKILL__

REPLACE ME: one paragraph on the job this skill does and the decision it supports.

## When it applies

- REPLACE ME: a concrete situation.

## When it does not apply

- REPLACE ME: the adjacent job that another skill or a human handles.

## Who does what

The model understands the request, searches and reasons. The repeatable artefact is produced by a deterministic script. If a step must return the same result on a rerun, it belongs in `scripts/`, not in prompt text.

| Step | Who runs it | Artefact |
|---|---|---|
| REPLACE ME | model / script / researcher | REPLACE ME |

## Procedure

1. REPLACE ME.
2. REPLACE ME.
3. REPLACE ME — the confirmation point: state exactly what the researcher has to approve before the work continues.

## Output format

REPLACE ME. Use `templates/report.md`.

## Quality gates

- Every claim traces to a source; what was not found is named as not found.
- Source coverage is shown explicitly, including access failures.
- Intermediate artefacts are kept, so the result can be rechecked rather than taken on trust.

## Boundaries

- REPLACE ME: what the skill does not guarantee (coverage of paywalled databases, legal assessment, authorship of conclusions).