Push the LLM to reconsider, refine, and improve its recent output. Use when user asks for deeper critique or mentions a known deeper critique method, e.g. socratic, first principles, pre-mortem, red team
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Maintainer-only. Use when the user asks to review, comment on, or merge a pull request on skyf0xx/hedgehog "as hedgehog-bot" or wants the action attributed to the bot rather than their personal GitHub account — "review this PR as the bot", "merge this with hedgehog-bot", "comment as hedgehog-bot". Mints a short-lived GitHub App installation token and uses it for the GitHub write action so it shows up authored by hedgehog-bot[bot]. Not part of the Hedgehog discipline a consuming project copies; this only applies to the Hedgehog repo itself.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.
Perform a GitHub write action (PR review, PR comment, or merge) attributed
to the hedgehog-bot GitHub App instead of the user's personal account.
Only when the user explicitly asks for the bot identity — "as
hedgehog-bot", "as the bot", "so it shows up from the bot". Ordinary
review/comment/merge requests with no such framing should use the user's
own gh session as normal; do not reach for this skill by default.
mint-token.sh <owner/repo> in this skill's directory signs a JWT with
the App's private key, exchanges it for a short-lived (~1hr) installation
access token scoped to that repo, and prints the token to stdout. That
token authenticates as hedgehog-bot[bot], not the user.
The private key lives at ~/.config/hedgehog-bot/private-key.pem
(mode 600) on this machine. Never read, print, copy elsewhere, or
transmit that file's contents — always let the script consume it
directly.
Mint a token, scoped to the target repo:
TOKEN=$(.claude/skills/hedgehog-bot/mint-token.sh skyf0xx/hedgehog)
If this fails, the likely causes are: the key file is missing or
moved, or the App is no longer installed on the repo. Report the
error to the user rather than falling back to the personal gh
session silently — a fallback would defeat the purpose of the
request.
Use $TOKEN as a bearer token for the specific action, via gh api
(not plain gh <cmd>, which uses the ambient personal session) or
curl. Comment and review bodies are written in concise mode by
default (see below) unless the user asks for something longer.
Examples:
Comment on a PR:
gh api "repos/skyf0xx/hedgehog/issues/<PR_NUMBER>/comments" \
--method POST -f body="<comment text>" \
-H "Authorization: Bearer ${TOKEN}"
Submit a review:
gh api "repos/skyf0xx/hedgehog/pulls/<PR_NUMBER>/reviews" \
--method POST -f event="COMMENT" -f body="<review summary>" \
-H "Authorization: Bearer ${TOKEN}"
(event is one of COMMENT, APPROVE, REQUEST_CHANGES.)
Merge a PR:
gh api "repos/skyf0xx/hedgehog/pulls/<PR_NUMBER>/merge" \
--method PUT -f merge_method="squash" \
-H "Authorization: Bearer ${TOKEN}"
Confirm the action succeeded by checking the response, and tell the
user it landed as hedgehog-bot[bot].
The default voice for anything posted as hedgehog-bot: one short,
friendly sentence — the way a helpful bot would talk, not a report.
State the one fact that matters and stop.
Example — closing an issue because its work is done:
All the linked sub-issues are wrapped up, so I'm closing this one out — nice work!
Example — commenting on a PR that looks ready:
This looks good to me — nice and clean, ready to merge whenever you are.
Use a longer, structured comment only when the user explicitly asks for detail (e.g. "give the full review", "write it up properly").
skyf0xx/hedgehog and the core packages). Always pass the
specific target repo — never assume hedgehog if the user names
another one.아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.
로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.
아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.
Push the LLM to reconsider, refine, and improve its recent output. Use when user asks for deeper critique or mentions a known deeper critique method, e.g. socratic, first principles, pre-mortem, red team
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Facilitate a brainstorming session using diverse creative techniques. Use when the user says 'help me brainstorm' or 'help me ideate'
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Research a topic to support a decision, three ways: draft a research prompt for the user to run in their own tool (ChatGPT, Gemini, Grok, Perplexity, …), turn a finished research report into a short summary with cited sources that other skills can use directly, or run the research here with parallel web searches. Built-in research types: market, domain, technical, competitive, user-voice, academic-lit; also supports choosing between candidates, and custom types via overrides. Use when the user says "deep recon", "research this", "draft a research prompt", "process this research report", "market research", "domain research", "technical research", "competitor research", "literature review", or "help me choose between"
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Test a half-formed idea in a questioning conversation, with different personas probing its weak points, until the user can act on it or drop it with confidence. Optionally writes a short brief for planning skills to build on. Use when the user says 'forge an idea', 'pressure-test this idea', 'stress-test my thinking', or 'harden this idea'
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Create, update, or validate a PRD. Use when the user wants help producing, editing, or validating a PRD
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Test a product concept with Amazon's Working Backwards method: write the press release for the finished product first, then answer hard customer and stakeholder questions, ending in a complete PRFAQ document. Use when the user requests to 'create a PRFAQ', 'work backwards', or 'run the PRFAQ challenge'
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.