Spec Research

What it does

When your agent uses it

• Tasks that involve Context engineering
• Tasks that involve Subagents

Example prompts

• “/spec-research”

Requirements

• Pre-approved tools (allowed-tools): Read, Grep, Glob, Bash(git:*), WebSearch, WebFetch

Workflow steps

6 steps, taken from the step headings in SKILL.md.

1. 1Frame the question
2. 2Map the local landscape first
3. 3External sources (in priority order)
4. 4Write research.md
5. 5Constraints on this artifact
6. 6Return to caller

What it can do on your machine

Read from SKILL.md and the folder at commit e5ee7e1. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves these tools, so the agent can use them without asking each time:

  • Read
  • Grep
  • Glob
  • Bash(git:*)
  • WebSearch
  • WebFetch

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown).

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Links to these hosts (documentation or services it may open):

  • github.com

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Spec Research loads about 1.1k tokens when it runs. Until then it costs about 73 tokens; SKILL.md has 290 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~73

When it runs · the whole SKILL.md, loaded when a task matches

~1.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from Smana/cloud-native-ref at commit e5ee7e1, republished under its Apache-2.0 licence (© Smana). 290 words, ~1,098 tokens.

name

spec-research

description

Research patterns, ecosystem tools, and best practices before writing a design. Runs in a forked Explore subagent so it can query Context7, WebSearch, and the whole codebase without consuming the main context window. Writes docs/superpowers/specs/YYYY-MM-DD-<slug>-research.md.

allowed-tools

Read, Grep, Glob, Bash(git:*), WebSearch, WebFetch

when_to_use

When the user says "research before designing", "what does the ecosystem say", "look at how others do this", "find existing patterns for X", "scan Context7"…

disable-model-invocation

true

argument-hint

<slug> "<research question>" — e.g. /spec-research valkey "Valkey caching composition best practices"

context

fork

agent

Explore

# Research: <one-line question>

**Topic**: <slug>
**Conducted**: <YYYY-MM-DD>
**Researcher**: Claude (spec-research subagent)

---

## TL;DR

<3–5 bullet points: the decisions the researcher would recommend and why>

## Standard stack

<Operators, libraries, images we should use. Pin to specific versions when docs list an LTS.>

| Component | Pick | Version | Source |
|-----------|------|---------|--------|

## Local patterns worth reusing

<File paths + 1-line note on what each shows. Copy existing patterns over invention.>

- `<path>`: <why it matters>

## Don't hand-roll

<Things the ecosystem already solves. Saves the spec author from designing these.>

## Common pitfalls

<Sharp edges from ecosystem docs / community. Especially: KCL mutation (#285),
Cilium bug #43493 for prefix delegation, anything that broke in this repo before.>

## Open questions surfaced

<Items to raise as open questions during brainstorming>

- [ ] <question>
- [ ] <question>

## References

- Context7: `<library-id>` <brief excerpt>
- <Vendor doc URL>
- Local files: `<path>`, `<path>`

© Smana, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .agents/skills/spec-research of Smana/cloud-native-ref.

Open the folder on GitHubat commit e5ee7e1