# Storytime Survey > This skill should be used when the user asks to "survey the codebase", "scan for artifacts", "what's in this repo", "inventory the docs", "run a survey", "check coverage", or wants to understand the codebase and existing documents without running a full storytime session. Produces a survey with coverage fingerprint. - Skill: `1ps0/storytime-survey` (Agent Skill) - Install (CLI): `npx skillmds@latest add 1ps0/storytime-survey` - Raw SKILL.md: https://api.skillmd.com/api/skills/1ps0/storytime-survey/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Coding & Dev Tools - Author: 1ps0 (https://skillmd.com/u/1ps0) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/1ps0/storytime-survey --- # Storytime Survey — Standalone Run a codebase survey and artifact scan without triggering a full storytime session. Produces a survey document with coverage fingerprint. ## Arguments Optional focus area or problem context: $ARGUMENTS ## Process ### 1. Check for Prior Surveys Look for existing `survey.md` files in `specs/.storytime/sessions/*/` or elsewhere in the repo. If found, read the coverage fingerprint and compute the delta: - Commit drift: `git rev-list ..HEAD` - Coverage gaps: paths unvisited in prior survey - Present to user: resurvey stale, extend to gaps, full resurvey, or trust prior ### 2. Codebase Scan Launch an Explore agent to survey the codebase: - If a focus area was provided, prioritize those paths - Identify: code structure, patterns, dependencies, constraints - Note: existing test coverage, build system, deployment setup ### 3. Artifact Scan Scan the entire repo for prior work artifacts: - Specs, docs, design records, agent definitions, team files - ADRs, RFCs, .kiro files, README files with design rationale - See the main storytime skill's `references/artifact-scan.md` for full scan targets and classification heuristics Classify each: **team-like**, **spec-like**, **config-like**, or noise. ### 4. Present Inventory Show the artifact inventory as a checklist: - Each item: checkbox, filename, short description (< 10 words if needed) - Tag each: `[team]` `[spec]` `[config]` - Pre-check items that seem relevant User can: - Toggle items, ask about specific files - Direct instruction ("ignore all the kiro stuff") - No further action needed — this is just the survey ### 5. Write Survey Document Write `specs/.storytime/sessions//survey.md` (storytime-native) or to the appropriate location for the repo's mode. Include: - Codebase context summary - Artifact inventory with classifications and user dispositions - **Coverage fingerprint** (REQUIRED): ```yaml fingerprint: commit: branch: paths_scanned: [...] paths_skipped: [...] paths_unvisited: [...] files_examined: files_total: coverage_ratio: <0.XX> artifacts_found: artifacts_classified: team: spec: config: ``` If no topic was established, use `general` or derive from the focus area. ## Output The survey document is the deliverable. The user can then: - Run `/storytime:storytime` with the survey as pre-existing context - Run `/storytime:consolidate` to organize discovered artifacts - Run `/storytime:absorb` to have a team interpret the findings - Do nothing — the survey stands alone as a codebase snapshot