haryy985/yunkun-console-preview/src-hunt-lane

Claude Code and agent skill by haryy985/yunkun-console-preview

Trust tier 3
  • community
Source check pending

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

Iterative authorized SRC discovery and validation lane. It keeps one bounded hypothesis active, prevents idle scanning, and returns every result through the common evidence contract.

Install

npx vskill@latest install haryy985/yunkun-console-preview/src-hunt-lane

SKILL.md

SRC Hunt Lane

Use this lane when the target is a confirmed authorized program or enterprise scope and the task requires structured discovery followed by focused validation. It is an orchestration lane, not a replacement for the API, business-flow, profiling, collection, or evidence lanes.

Mode Selection

InputMode
Fixed domains, hosts, or applicationsbounded scope mode
Authorized program with a target family but incomplete inventorydiscovery mode
Source tree or build artifactwhite-box mode

Do not widen scope from a weak technology match. A new asset needs an independent scope signal or explicit authorization.

Closed Loop

  1. Confirm scope, identities, request limits, exclusions, and stop conditions.
  2. Create or update the target profile.
  3. Seed passive collection from the approved target set.
  4. Normalize and deduplicate assets, routes, scripts, and parameters.
  5. Pick one seed with the strongest combination of reachability, privilege,

state impact, and evidence quality.

  1. Build one hypothesis with a control request and one expected invariant.
  2. Route to exactly one specialist lane.
  3. Validate the smallest reversible change.
  4. Record confirmed, incomplete, negative, or blocked.
  5. Return to the queue only after cleanup and evidence recording.

Anti-Idle Gates

  • A route name, banner, status code, or scanner hit is a lead only.
  • Two consecutive no-change checks require a hypothesis review.
  • Repeated authorization failures require a route change or a stop record.
  • Rate limits pause the source and do not justify source rotation.
  • Do not run a full matrix before the current seed has a decision.
  • Do not enumerate unrelated objects or identities.

Signal Routing

SignalRoute
HTML, JavaScript, API routes, parameters, or rendered statesnoweyes-collector
Asset graph, technology, host, or trust boundarytarget-profiling
API, token, object, schema, or transport boundaryapi-security
Payment, order, quota, approval, workflow, or state integritybusiness-logic-attack
SQL injection signalsqli-sql-injection
Reproduction, severity, or submission decisionsrc-report-paradigm

Canonical detailed checks live under skills/skill/knowledge-base/ and references/playbooks/. The expert modules under references/experts/ are optional depth references selected by the primary lane.

Seed Record

text
seed_id:
scope:
asset:
source:
collected_at:
route_family:
identity:
object_or_tenant:
signal:
hypothesis:
control:
mutation:
stop_condition:
status:
next_action:

Handoff Rules

Every specialist result must include the seed ID, the original source, the control, the mutation, the response delta, the state effect, cleanup, and evidence references. If the specialist cannot decide safely, return blocked with the exact missing prerequisite.

This lane is complete only when the current seed is closed, handed to the evidence lane, or explicitly stopped with a reason.

Security

  • Security scanPassed, 95 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

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

Source

Popularity

7-day trend score
0

Not refreshed yet.

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/haryy985/yunkun-console-preview/src-hunt-lane/badge)](https://verified-skill.com/skills/haryy985/yunkun-console-preview/src-hunt-lane)

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