Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/marcuskrogh-explore
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: research
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/explore
What's inside
- Skill name
explore- Description
- Exploration through foggy or oversized work. Charts ROADMAP.md plus a Story of sequenced route Tasks (one delivery unit by default). Research, model, and sandbox artifacts stay on the delivery branch — no separate PRs. Use when the destination is felt but the route is not yet visible.
- License
- MIT
- Instructions
- 152 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Explore Applies [CONCEPT_ALIGNMENT](../concepts/CONCEPT_ALIGNMENT.md) and [CONCEPT_SKILL](../concepts/CONCEPT_SKILL.md) to **wayfinding** — charting a course through foggy work. Explore charts. The explore workflow walks. **On invoke:** read [../concepts/CONCEPT_ALIGNMENT.md](../concepts/CONCEPT_ALIGNMENT.md), [CONCEPT_SKILL](../concepts/CONCEPT_SKILL.md), and [../workflow/SKILL.md](../workflow/SKILL.md). ## Leading words - **delivery unit** — one define-typed Task (and its PR) that owns research/model/sandbox artifacts through ship when they share a build - **finding docs** — `RESEARCH.md` / `MODEL.md` committed on the delivery branch for later skills; never their own PR - **supportive-only** — explore route Task that does not itself implement or ship; its artifacts sit on the delivery branch **Chart the route.** Creating a research/model/sandbox/define Task completes a map step. The explore workflow reads the frontier type. ## Extensions | Slot | This skill | |------|------------| | **Subject** | Vague, foggy, or oversized work | | **Probes** | Destination (1–2 lines); why it matters; competing framings that change the whole map; fog vs ticket; route step kinds; sequence/deps; out of scope | | **Stop condition** | Destination named, visible frontier ticketed (with sequence/deps), remaining uncertainty recorded as fog or out of scope | | **Alignment artifact** | `ROADMAP.md` (path from WORKSPACE) — the map | | **Readiness prompt** | "Does this map capture the destination and the next steps through the fog?" | | **Opening** | Thin: "What are we trying to find our way to?" Rich: highest-value fog (destination, framing, or first takeable step). Existing map → [Continue the map](#continue-the-map) | | **Scope guard** | Chart only — no destination implementation; no fake-settled requirements for later skills; tracker writes after approval are charting, not walking; no map-only open PR | **Divergence here** changes destination, route shape, or Task existence/deps. Choices that only change how a later skill answers belong on that Task — park a one-line fog/Task pointer. ## Fog of war **Fog or ticket?** Can you state the step precisely enough to act on *now* — not whether you can answer it now. | Put it here | When | |-------------|------| | **Route Task** | Sharp enough to ticket as type define (preferred), research, model, sandbox, or task — even if blocked | | **Not yet specified** (fog) | Sensed but not ticket-sized — do not pre-slice into fake Tasks | | **Out of scope** | Beyond this destination — never graduates unless destination is redrawn | Resolving a route Task should clear fog ahead. Re-invoking `/explore` **graduates** what is now sharp into fresh Tasks. ## Route Task types | Type | Role | |------|------| | **research** | Evidence finding docs the route waits on | | **model** | Math finding docs before (or with) definition | | **sandbox** | Isolated inspect-loop for a contained element before production implement | | **define** | User-agent alignment on particulars for a buildable slice | | **task** | Other unblocker that earns its place by unblocking a later decision | **One delivery unit by default.** Prefer a **single define-typed route Task** when research / model / sandbox / define feed the same eventual build — run those steps on that Task so artifacts land on the same branch define and implement will use. Prefer **separate route Tasks with dependencies** only when sequence truly needs independent tickets (parallel owners, research that may kill the destination, or blockers owned elsewhere). Wire **Blocked by** after Tasks have keys. A Task is **unblocked** when every blocker is **Done** (or waived). The frontier is the unblocked row with the lowest Order. The explore workflow maps its type to a skill. **Finding docs, not separate PRs.** Research and model only produce finding documentation on the delivery branch. Sandbox produces an isolation tree and `SANDBOX.md` the same way. They never open their own PRs. Explore does not open a map-only PR. The only open delivery PR is the delivery head once a definition skill opens it. Supportive-only route Tasks go **Done** after their docs (or isolation tree) are on the downstream delivery branch. ## Artifact `ROADMAP.md` is an **index**; detail lives on tickets. ```markdown # Roadmap: [title] ## Destination <one or two lines> ## Notes <domain hints; standing preferences> ## Route | Order | Task | Type | Blocked by | Status | Issue | |-------|------|------|------------|--------|-------| | 1 | <title> | research / model / sandbox / define / task | — | To Do | <KEY> | ## Cleared so far - [<title>](link) — <one-line gist> ## Not yet specified - … ## Out of scope - … ## Tracker - Provider: … - Story (map): <KEY> - Tasks: <KEY>, … ``` Keep Destination short. **Not yet specified** stays non-empty whenever honest fog remains. ## Steps ### Chart the map 1. **Align** — Follow CONCEPT_ALIGNMENT with the extensions above. Done when the destination and map stop condition hold. 2. **Classify the route** — Map the visible frontier breadth-first into typed Tasks, fog, and out of scope. Done when every visible item has exactly one classification and dependencies are known. 3. **Handle a clear direct path** — If the work is already small and fog-free, record a one-row route of type **define** instead of manufacturing extra Tasks. Done when that row is the frontier.
From marcuskrogh/skills, MIT.
More from marcuskrogh/skills
- 0.587score
define
marcuskrogh
Definition front door for concrete work.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
help
marcuskrogh
Help choosing skills and workflows.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
iterate
marcuskrogh
Iteration on already merged work.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
adopt
marcuskrogh
Adopt the structure catalog across an existing codebase that was not built to it.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
implement
marcuskrogh
Implementation through managed, value-routed work packages.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
refine
marcuskrogh
Refine alignment and lightweight definition for a limited area that needs refactoring or descriptive improvement without changing behaviour.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
rework
marcuskrogh
Rework alignment and lightweight definition for an intentional implementation change that must not degrade measured outcomes.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
model
marcuskrogh
Mathematical alignment through LaTeX-only questions.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
restructure
marcuskrogh
Restructure: refactoring pass against the structure catalog (campground, names, size, cohesion, seams, named smells, CRAP).
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
review
marcuskrogh
Review of a change as sequential or bundled lasers across Spec, Correctness, Integration, Architecture, and Standards: find, fix, then publish a code review.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
architect
marcuskrogh
Architecture for one Task: record where the work sits (modules, layers, seams, dependency direction) in ARCHITECTURE.md.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
bug
marcuskrogh
Bug alignment and lightweight definition for a clear defect.
SkillAFreshness: A★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 1
- Open issues
- 0
- Last commit (this folder)
- Sep 27, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/marcuskrogh/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from marcuskrogh
- 0.587score
iterate
marcuskrogh
Iteration on already merged work.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
help
marcuskrogh
Help choosing skills and workflows.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
adopt
marcuskrogh
Adopt the structure catalog across an existing codebase that was not built to it.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
define
marcuskrogh
Definition front door for concrete work.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
rework
marcuskrogh
Rework alignment and lightweight definition for an intentional implementation change that must not degrade measured outcomes.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
refine
marcuskrogh
Refine alignment and lightweight definition for a limited area that needs refactoring or descriptive improvement without changing behaviour.
SkillAFreshness: A★ 1Claude Code +9
Cite this page
explore skill. Agent Mart. https://agentmart.dev/skills/marcuskrogh-explore (last updated 2026-10-24)