How do I install the Vaults skill?

Run npx skillmds@latest add majiayu000/vaults in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Vaults safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Vaults?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Vaults free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Vaults?

majiayu000 (@majiayu000) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Vaults by majiayu000 — Agent Skill

Name: Vaults
Rating: 4.666666666666667 (3 reviews)
Author: majiayu000

Vaults

Manages case.dev encrypted document vaults for legal workflows. Creates vaults, uploads files and directories, lists and downloads objects, and runs semantic search across vault contents. Use when the user mentions "vault", "upload documents", "document storage", "download files", "vault search", or needs to manage case files and discovery documents.

majiayu000 Updated Sep 9, 2026 567 repo stars

File contents

case.dev Vaults

Encrypted document storage with automatic OCR, chunking, and semantic search. Each vault is an isolated container for a case, project, or document collection.

Requires the casedev CLI. See setup skill for installation and auth.

Create a Vault

casedev vault create --name "Smith v Jones Discovery" --description "Phase 1 documents" --json

Flags:

--name / -n (required)
--description / -d
--disable-graph — skip knowledge graph construction
--disable-indexing — skip automatic ingestion (upload-only mode)

List Vaults

casedev vault list --json
casedev vault list --wide --json   # includes description column

Get Vault Details

casedev vault get VAULT_ID --json

Also accepts focused vault if set via casedev focus set --vault VAULT_ID.

Upload Files

Single file

casedev vault object upload ./contract.pdf --vault VAULT_ID --json

Flags: --name (override filename), --content-type, --no-ingest (skip indexing).

Directory (recursive)

casedev vault upload ./discovery-docs/ --vault VAULT_ID --json

Preserves folder structure. Each file gets a presigned URL upload with confirmation. Use --no-ingest to upload without triggering indexing. Use --content-type to force a MIME type for all files.

List Objects in a Vault

casedev vault object list --vault VAULT_ID --json

Returns: object ID, filename, ingestion status, size.

Download Files

# Download a single object
casedev vault download --vault VAULT_ID --object OBJECT_ID --out ./output/ --json

# Download all objects
casedev vault download --vault VAULT_ID --out ./all-docs/ --json

# Download by path prefix
casedev vault download --vault VAULT_ID --path "exhibits/" --out ./exhibits/ --json

Semantic Search

casedev search vault "force majeure clause" --vault VAULT_ID --json

Search methods:

--method hybrid (default) — combines vector + keyword search
--method fast — keyword-only, faster
--method global — cross-vault if supported

Additional flags:

--limit / -l — max results (default 10)
--object OBJECT_ID — restrict search to specific object(s), repeatable

Returns: answer summary, source documents, matching chunks with page ranges and scores.

If hybrid mode returns a server error, retry with --method fast.

Common Workflows

Upload and search a case file set

# 1. Create vault
casedev vault create --name "Acme v Beta" --json

# 2. Upload documents
casedev vault upload ./case-files/ --vault VAULT_ID --json

# 3. Wait for ingestion (check object status)
casedev vault object list --vault VAULT_ID --json

# 4. Search
casedev search vault "indemnification obligations" --vault VAULT_ID --json

Focus workflow (skip --vault on every command)

casedev focus set --vault VAULT_ID
casedev vault object list --json
casedev search vault "damages calculation" --json
casedev vault download --object OBJ_ID --out ./downloads/ --json

Troubleshooting

"Missing vault ID": Provide --vault VAULT_ID or set focus with casedev focus set --vault VAULT_ID.

Upload fails with HTTP 4xx: Check file size (max 500MB per object), verify vault ID exists, confirm auth.

Search returns no results: Ensure objects have completed ingestion (status: "completed" in object list). Try --method fast if hybrid fails.

---name: vaultsdescription: Manages case.dev encrypted document vaults for legal workflows. Creates vaults, uploads files and directories, lists and downloads objects, and runs semantic search across vault contents. Use when the user mentions "vault", "upload documents", "document storage", "download files", "vault search", or needs to manage case files and discovery documents.--- # case.dev Vaults Encrypted document storage with automatic OCR, chunking, and semantic search. Each vault is an isolated container for a case, project, or document collection. Requires the `casedev` CLI. See `setup` skill for installation and auth. ## Create a Vault ```bash casedev vault create --name "Smith v Jones Discovery" --description "Phase 1 documents" --json ``` Flags: - `--name` / `-n` (required) - `--description` / `-d` - `--disable-graph` — skip knowledge graph construction - `--disable-indexing` — skip automatic ingestion (upload-only mode) ## List Vaults ```bash casedev vault list --json casedev vault list --wide --json # includes description column ``` ## Get Vault Details ```bash casedev vault get VAULT_ID --json ``` Also accepts focused vault if set via `casedev focus set --vault VAULT_ID`. ## Upload Files ### Single file ```bash casedev vault object upload ./contract.pdf --vault VAULT_ID --json ``` Flags: `--name` (override filename), `--content-type`, `--no-ingest` (skip indexing). ### Directory (recursive) ```bash casedev vault upload ./discovery-docs/ --vault VAULT_ID --json ``` Preserves folder structure. Each file gets a presigned URL upload with confirmation. Use `--no-ingest` to upload without triggering indexing. Use `--content-type` to force a MIME type for all files. ## List Objects in a Vault ```bash casedev vault object list --vault VAULT_ID --json ``` Returns: object ID, filename, ingestion status, size. ## Download Files ```bash # Download a single object casedev vault download --vault VAULT_ID --object OBJECT_ID --out ./output/ --json # Download all objects casedev vault download --vault VAULT_ID --out ./all-docs/ --json # Download by path prefix casedev vault download --vault VAULT_ID --path "exhibits/" --out ./exhibits/ --json ``` ## Semantic Search ```bash casedev search vault "force majeure clause" --vault VAULT_ID --json ``` Search methods: - `--method hybrid` (default) — combines vector + keyword search - `--method fast` — keyword-only, faster - `--method global` — cross-vault if supported Additional flags: - `--limit` / `-l` — max results (default 10) - `--object OBJECT_ID` — restrict search to specific object(s), repeatable Returns: answer summary, source documents, matching chunks with page ranges and scores. If hybrid mode returns a server error, retry with `--method fast`. ## Common Workflows ### Upload and search a case file set ```bash # 1. Create vault casedev vault create --name "Acme v Beta" --json # 2. Upload documents casedev vault upload ./case-files/ --vault VAULT_ID --json # 3. Wait for ingestion (check object status) casedev vault object list --vault VAULT_ID --json # 4. Search casedev search vault "indemnification obligations" --vault VAULT_ID --json ``` ### Focus workflow (skip --vault on every command) ```bash casedev focus set --vault VAULT_ID casedev vault object list --json casedev search vault "damages calculation" --json casedev vault download --object OBJ_ID --out ./downloads/ --json ``` ## Troubleshooting **"Missing vault ID"**: Provide `--vault VAULT_ID` or set focus with `casedev focus set --vault VAULT_ID`. **Upload fails with HTTP 4xx**: Check file size (max 500MB per object), verify vault ID exists, confirm auth. **Search returns no results**: Ensure objects have completed ingestion (status: "completed" in object list). Try `--method fast` if hybrid fails.

Frequently asked questions