backbay-labs/thrunt-god/hunt-new-case

Claude Code and agent skill by backbay-labs/thrunt-god

Trust tier 2
  • community
LLM review pending
Source check pending

Checking that the SKILL.md file is still in the repository.

Initialize a threat hunting case from a signal, detection, intel lead, or analyst suspicion

Install

npx vskill@latest install backbay-labs/thrunt-god/hunt-new-case

SKILL.md

The same SKILL.md was published earlier as backbay-labs/thrunt-god/hunt-new-case.

<context> Flags:

  • --auto - Use the supplied signal brief as the starting point and ask only for missing critical facts.
  • --pack <id> - Bootstrap the case from a built-in or local hunt pack. Use thrunt-tools pack bootstrap <id> to inspect the generated mission, hypothesis, and phase seed content.

</context>

<objective> Initialize a threat hunting case.

These hunt-native artifacts are the source of truth for the case.

Creates:

  • .planning/config.json
  • .planning/MISSION.md
  • .planning/HYPOTHESES.md
  • .planning/SUCCESS_CRITERIA.md
  • .planning/HUNTMAP.md
  • .planning/STATE.md
  • .planning/QUERIES/
  • .planning/RECEIPTS/

Bootstrap should only scaffold the case. Do not seed sample queries, sample receipts, or completed phases. Unknown scope details, data sources, operators, and constraints must remain TBD unless the operator confirms them. Confirmed bootstrap facts such as the case name, mode, opened date, and initial phase/status must be filled immediately.

After this command: Run /hunt-shape-hypothesis or /hunt-plan 1. </objective>

<execution_context> @.github/thrunt-god/workflows/hunt-bootstrap.md @.github/thrunt-god/templates/config.json @.github/thrunt-god/templates/mission.md @.github/thrunt-god/templates/hypotheses.md @.github/thrunt-god/templates/success-criteria.md @.github/thrunt-god/templates/huntmap.md @.github/thrunt-god/templates/hunt-state.md </execution_context>

<process> Execute the bootstrap workflow from @.github/thrunt-god/workflows/hunt-bootstrap.md in case mode. Focus on turning the input signal into a scoped case with explicit hypotheses, data sources, and evidence requirements. When --pack <id> is present, use the pack bootstrap output as the default case skeleton and ask only for the missing pack parameters or signal-specific overrides. Create .planning/QUERIES/ and .planning/RECEIPTS/ as empty directories only. Do not load query-log or receipt templates during bootstrap; those belong to /hunt-run after real execution begins. Default behavior is scaffold-first: write confirmed facts only and leave unknown values as TBD instead of inventing sample content. Create .planning/config.json during bootstrap if it does not already exist so runtime, settings, and connector commands are immediately usable. Never hand-write .planning/config.json; use thrunt-tools config-new-program and thrunt-tools config-set so the file stays valid THRUNT config. Use built-in connector ids exactly as the runtime registers them, for example splunk and elastic; do not substitute elasticsearch. When writing connector profiles, use base_url for the runtime URL field; do not invent or substitute endpoint. Only configure connector profiles when auth type and secret ref names are confirmed. Never invent placeholder env vars or placeholder secrets for blocked connectors. When writing secret_refs, each confirmed secret must use the THRUNT object shape { "type": "env", "value": "ENV_VAR_NAME" } rather than a raw string. Keep connector narrative, status notes, and access commentary in ENVIRONMENT.md, not in ad hoc config keys. Do not leave bootstrap-known fields as TBD after writing the files. Write the hunt artifacts directly. </process>

Security

  • Security scanPassed, 100 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
30
Repository forks
5
Repository npm downloads, 7 days
15
7-day trend score
33

Stars and forks are for the whole GitHub repository, not this skill alone. Downloads are for the repository's npm package. Refreshed 30 Apr 2026, 6 months ago.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/backbay-labs/thrunt-god/hunt-new-case/badge)](https://verified-skill.com/skills/backbay-labs/thrunt-god/hunt-new-case)

Something wrong with this skill or its page? Tell us.