agentsclimarketplace

Vaults

Skill CaseMark/legal-plugin/vaults

Claude Skills for case.dev — legal AI vaults, OCR, transcription, and search

Install
npx -y skills add CaseMark/legal-plugin --skill vaults

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

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.

SKILL.md

3.7 KB, 873 tokens by cl100k_base, as published. Nobody here has run it

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.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.