# Research Only

> Use this skill ALWAYS when the user asks "разберись", "изучи", "посмотри что это", "что ты думаешь о...", "what do you think", "investigate", "explore", or any phrase WITHOUT an explicit action verb. Trigger this skill EVEN if it seems like a small question — if no action verb (правь/делай/фикси/добавь) is present, mode is research and outputs only text, no code edits, no file writes, no commits.

- Skill: `engsimsoft/research-only` (Agent Skill)
- Install (CLI): `npx skillmds@latest add engsimsoft/research-only`
- Raw SKILL.md: https://api.skillmd.com/api/skills/engsimsoft/research-only/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: engsimsoft (https://skillmd.com/u/engsimsoft)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/engsimsoft/research-only

---


# Research-only mode (R17 strict enforcement)

When user message has NO action verb (правь, делай, фикси, добавь, реализуй, fix, build, implement, edit) — you are in RESEARCH mode.

## What you DO
- Read code via Read, Glob, Grep, Serena `find_symbol`, `get_symbols_overview`
- Fetch external docs via Context7 MCP, WebFetch, WebSearch
- Use Explore subagent for wide search (delegates to fresh context, doesn't pollute main)
- Synthesize findings into structured text answer

## What you DO NOT
- ✗ Edit any file
- ✗ Write any file
- ✗ Run state-changing commands (git, npm install, alembic, etc.)
- ✗ Make commits
- ✗ Modify git state

## Format of answer
```
Short summary (2-3 sentences):
...

Key findings (bulleted):
- Finding 1 (file:line)
- Finding 2 (file:line)
...

Proposed next steps (R4 — propose, don't ask):
- I would do X first because Y. Stop me if different direction.
- Then Z. Then W.

If you want me to do <X> — say so explicitly.
```

## When research uncovers serious problem
- DO NOT secretly fix it — that breaks research-only contract
- DO record finding in `docs/agent/06_known_issues.md` if explicitly broken
- DO propose fix in your text answer, but do not execute

## Subagent delegation
For wide research (>5 files, >3 web searches) — delegate via Agent tool with subagent_type="Explore":
- Subagent gets isolated context, returns ≤200-word summary
- Main context stays clean (R7 budget)
- This is mandatory for repo-wide audits, not optimization

## Anti-patterns
- ✗ "Research" the bug, then quietly fix it without permission
- ✗ Ramble for 1000 words because "research = thorough"
- ✗ End with "Should I do X?" → use R4 propose format instead