Skip to content

Kartik-Doye/Ravengard-AI-Recruiter / .agents/skills/spelunk

spelunk

Verified

Maps an unfamiliar repository from a concrete question by finding manifests, entry points, execution paths, dependencies, churn, ownership, and risk. Use for codebase onboarding, impact analysis, or locating where behavior lives before planning or debugging.

FileKindSize
SKILL.mdskill md9.5 KB
references/discovery-map.mdmarkdown3.7 KB
references/evidence-protocol.mdmarkdown1.5 KB
references/research.mdmarkdown768 B
scripts/inventoryother4.4 KB

references/evidence-protocol.md

View on GitHub
# Discovery evidence protocol

Use this for deep maps, risk maps, monorepos, or work that a later planner or
debugger will rely on. Quick answers need only the relevant observed path.

## Evidence record

| Field | Capture |
| --- | --- |
| Question and target | decision to support and app/package boundary |
| Provenance | repository root, commit or tree state, and check date |
| Observed paths | manifests, entrypoints, configuration, symbols, tests read |
| Traced path | one real route, command, or lifecycle with symbol anchors |
| Inferences | conclusion plus observed evidence supporting it |
| Unknowns | missing runtime, ownership, generated, or external evidence |
| Recheck trigger | commit, package, config, or task change that invalidates map |

## Monorepo routing

Identify workspace manager, package graph, target app, shared packages, and
deployment ownership. Follow only dependencies that change the target behavior.
Report sibling packages as not covered unless read; do not generalize one app's
scripts or patterns to the whole repository.

## Risk-map evidence

A hotspot needs an observed complexity signal, churn signal when Git is
available, trust boundary, test gap, or contradiction. Churn alone is not a
defect. Name the likely owning skill and the evidence gap before recommending a
change.

## Handoff rule

Give next worker source paths, symbols, command definitions, active unknowns,
and the next narrow read. Do not paste inventory output or turn a map into an
implementation plan.
planning-and-task-breakdownaddyosmani/agent-skillsBreaks work into ordered tasks. Use when you have a spec or clear requirements and need to break work into implementable tasks. Use when a task feels too large to start, when you need to estimate scope, or when parallel work is possible.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
screenpipethedotmack/claude-memSearch screen and audio history recorded by Screenpipe and optionally save selected context into claude-mem. Use when the user asks to recall desktop activity, a meeting, or something seen in another app through Screenpipe.Stars: 118kUpdated Claude CodeCodexVerifiedInstall
contact-widgetnexu-io/open-designSelf-contained floating chat widget with welcome screen, social links, meeting button, and message input. Single HTML file, zero dependencies.Stars: 108kUpdated Claude CodeCodexVerifiedInstall
meeting-notesnexu-io/open-designMeeting notes page — title bar with attendees, agenda checklist, decisions block, action items table with owners + dates, and a "next meeting" footer. Use when the brief mentions "meeting notes", "minutes", "1:1 notes", "all-hands recap", or "会议纪要".Stars: 108kUpdated Claude CodeCodexVerifiedInstall