prompts/skills/research/SKILL.md
Pre-implementation research for design, architecture, and impact analysis. Use during brainstorming, architectural design, or before implementation to understand dependencies and integration points. Triggers on (1) designing or architecting a feature, (2) brainstorming implementation approaches, (3) analyzing impact of proposed changes, (4) exploring reference materials in extra/, (5) creating research reports before planning.
npx skillsauth add ramblurr/nix-devenv researchInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
extra/ for required reference materials. HALT if missing.extra/.prompts/NNN-concept_report.md.All research relies on materials in the extra/ directory.
This includes reference codebases, documentation, and specifications.
Before proceeding with any research:
extra/Never use web fetch or web search as the primary research method.
The extra/ directory should contain everything needed.
If required documentation or codebase is not in extra/, you must:
Example HALT message:
HALT: Missing reference material.
Required: Textual compositor implementation
Suggestion: Clone to extra/textual:
git clone https://github.com/Textualize/textual extra/textual
Required: Ratatui border symbols
Location needed: extra/ratatui/ratatui-core/src/symbols/
Suggestion: If not present, clone:
git clone https://github.com/ratatui/ratatui extra/ratatui
Do not attempt to work around missing materials. Do not use web search as a substitute. Wait for the user to add the required materials.
prompts/bd create "<concept>" -t epic --from-template epic --json
bd create "Research: <concept>" -t task --from-template research --parent <epic-id> --json
bd close <research-task-id> --reason "Research complete: prompts/NNN-concept_report.md"
Discover what the project already uses:
Never:
These agents are READ-ONLY - they return information and suggestions only. You are the orchestrator: guide these agents, synthesize their outputs into the final research document.
| Agent | When to Consult | What They Return |
|-------|-----------------|------------------|
| Explore | For codebase exploration and search | File locations, code patterns, structural understanding |
| clojure-expert | For Clojure patterns and idioms | Idiomatic approaches, style guidance |
| research-agent | For analyzing extra/ materials | Extracted patterns, API summaries |
Use Explore agents extensively to:
List materials from extra/ that were analyzed:
extra/textual/src/textual/_compositor.py - Compositor implementationextra/ratatui/ratatui-widgets/src/borders.rs - Border symbolsFrom deps.edn:
Document patterns already in use:
src/ol/bramble/widgets/text.clj:15src/ol/bramble/protocols.clj:20path/to/file.clj:line - [specific change needed]path/to/file.clj:line - [specific change needed]Questions for the user:
You have READ-ONLY git access.
Allowed:
git statusgit loggit diffgit showNever allowed:
git add, git commit, git pushSave research to: prompts/NNN-concept_report.md
Or for specific library research: prompts/NNN-concept_library_report.md
Check existing files in prompts/ to determine the next NNN number.
Follow the naming convention: three digits, hyphenated concept label, underscore report suffix.
After completing research, output one of the following:
## Research Complete
Document: prompts/NNN-concept_report.md
Ready for implementation planning.
testing
Use this OCP when executing or preparing to execute commands that change a live or important system, service reloads/restarts, package changes, deployments, migrations, firewall/network/access changes, credential rotation, NixOS switch/test/boot/deploy, or incident mitigation. It guides safe operations with a persisted ledger for scope, preflight, baseline, rollback, validation, and evidence.
development
Create new agent skills with proper structure, progressive disclosure, and bundled resources. Use when user wants to create, write, or build a new skill.
documentation
Naming conventions for workflow documents in prompts/. Use when creating plans, PRDs, research reports, idea capture or other workflow documents. Triggers on (1) creating new planning documents, (2) naming PRDs or research reports, (3) questions about document organization in prompts/.
testing
Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when user wants to stress-test a plan against their project's language and documented decisions.