Skip to content
Back to skills

Cargo Content

ASecurity

Manage the knowledge a Cargo workspace holds — upload files (PDF, CSV, text), rename and organize them, and build native or connector-backed libraries that sync from an external source, so agents can retrieve them (RAG). Triggers: \"upload this PDF\", \"add these docs as knowledge\", \"build a knowledge base\", \"sync our help center into Cargo\", \"what files are in the workspace\", \"index this folder\", \"attach our pricing sheet\". Skip when: wiring the file or library into an agent — use...

  • 19 stars
  • 0 votes
  • 0 copies
  • 4 views
  • Added August 31, 2026
ai-agentsgobashnodegitapi

Works with

  • cli
  • api

Security analysis

A96/100
  • mediumInstalls packages at runtime which could introduce malicious dependencies

Pro scans all 5 files and shows the line behind each finding

Scanned August 31, 2026

npx -y skills add getcargohq/cargo-skills --skill cargo-content --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Cargo Content?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Cargo Content
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/getcargohq-cargo-content/badge)](https://www.skillsdirectory.com/skills/getcargohq-cargo-content)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: cargo-content
description: "Manage the knowledge a Cargo workspace holds — upload files (PDF, CSV, text), rename and organize them, and build native or connector-backed libraries that sync from an external source, so agents can retrieve them (RAG). Triggers: \"upload this PDF\", \"add these docs as knowledge\", \"build a knowledge base\", \"sync our help center into Cargo\", \"what files are in the workspace\", \"index this folder\", \"attach our pricing sheet\". Skip when: wiring the file or library into an agent — use cargo-ai; uploading a CSV that drives a batch run — that is a workspace-management file upload."
version: "1.0.1"
compatibility: Requires @cargo-ai/cli (npm). Sign in or create an account with `cargo-ai login --email` (emailed code, no browser), `--oauth`, or an API token
homepage: https://github.com/getcargohq/cargo-skills
metadata:
  author: getcargo
  openclaw:
    requires:
      bins:
        - cargo-ai
    install:
      - kind: node
        package: "@cargo-ai/cli@latest"
        bins:
          - cargo-ai
    homepage: https://github.com/getcargohq/cargo-skills
---

# Cargo CLI — Content

Workspace **knowledge** management: upload and organize **files** (PDFs, CSVs, text) and group or sync them into **libraries**. These are the binary/grouped knowledge resources that ground agent responses (retrieval-augmented generation, RAG).

> **New top-level domain (CLI ≥ 1.0.19).** Files and libraries moved out of the `ai` domain into the `content` domain — `cargo-ai content file …` and `cargo-ai content library …`. The old `cargo-ai ai file …` commands no longer exist; an `unknown command` error means you're on the old path.

> For **attaching** a file or library to an agent (via the release `resources` array), use [`cargo-ai`](../cargo-ai/SKILL.md).
> For **folders** that organize files, use [`cargo-workspace-management`](../cargo-workspace-management/SKILL.md) (`cargo-ai workspaceManagement folder …`).
> For batch-run **input** files (CSVs uploaded to drive a batch), that's a different surface — `cargo-ai workspaceManagement file upload` — documented in [`cargo-workspace-management`](../cargo-workspace-management/SKILL.md).

> See `references/examples/files.md` for end-to-end file, library, and attach-to-agent examples.
> See `references/response-shapes.md` for full JSON response structures.
> See `references/troubleshooting.md` for common errors and how to fix them.

## Bootstrap

Already signed in (`cargo-ai whoami` returns a workspace)? Skip to the next section.

```bash
npm install -g @cargo-ai/cli            # no global install? prefix every command with `npx @cargo-ai/cli`
cargo-ai login --email you@company.com  # emailed code, no browser; creates the account on first use
                                        # alternatives: --oauth (browser) · --token <api-token> (CI)
cargo-ai whoami                         # confirm the active workspace before any write
```

Every command prints JSON to stdout; failures exit non-zero with `{"errorMessage": "..."}`. Anything that creates a run or a batch is async — pass `--wait-until-finished` or poll the matching `get`. When the full skill bundle is installed, [`../cargo/references/prerequisites.md`](../cargo/references/prerequisites.md) adds the CLI version pin, token scopes, and the admin-only surface.

## Discover resources first

```bash
cargo-ai content file list                 # all uploaded files (uuid, name, contentType, size)
cargo-ai content library list              # knowledge libraries (native or connector-backed)
```

## Files

Upload files (PDFs, CSVs, text) so an agent can ground its responses in specific knowledge. The upload response includes the `uuid` you reference when attaching the file to an agent release (see [`cargo-ai`](../cargo-ai/SKILL.md)).

```bash
# List all files
cargo-ai content file list

# Get a single file
cargo-ai content file get <file-uuid>

# Upload a file (optionally straight into a folder)
cargo-ai content file upload --file ./knowledge-base.pdf
cargo-ai content file upload --file ./knowledge-base.pdf --folder-uuid <folder-uuid>

# Update a file's name or folder
cargo-ai content file update --uuid <file-uuid> --name "Q1 Research Notes"
cargo-ai content file update --uuid <file-uuid> --folder-uuid <folder-uuid>

# Remove a file
cargo-ai content file remove <file-uuid>
```

> **Reading content files from the context sandbox.** Uploaded content files are also available **read-only** under `.files/` in the context runtime sandbox, so a command run there can consume them — e.g. `cargo-ai context runtime execute --command ls --args '["-1",".files"]'`. The directory sits outside the committed context tree (never pushed, not writable); to add or change files use `content file` here. See [`cargo-context`](../cargo-context/SKILL.md).

## Libraries

A library groups files into one resource an agent can reference. There are two kinds:

- **`native`** — workspace-managed collections of uploaded files.
- **`connector`** — synced from an external source (e.g. a help center or knowledge base) through an unstructured-data extractor (`--extractor-slug`). Get the `connectorUuid` from [`cargo-connection`](../cargo-connection/SKILL.md).

```bash
# List libraries (filter by kind or connector)
cargo-ai content library list
cargo-ai content library list --kind native
cargo-ai content library list --kind connector --connector-uuid <connector-uuid>

# Get a single library
cargo-ai content library get <library-uuid>

# Create a connector-backed library
cargo-ai content library create \
  --name "Help Center" \
  --connector-uuid <connector-uuid> \
  --extractor-slug <extractor-slug> \
  --folder-uuid <folder-uuid> \
  --config '{}'

# Update / remove
cargo-ai content library update --uuid <library-uuid> --name "Updated Name"
cargo-ai content library remove <library-uuid>
```

## Attaching to an agent

Files and libraries are knowledge **resources** — they do nothing until attached to an agent via that agent's draft release `resources` array, then deployed. That wiring lives in [`cargo-ai`](../cargo-ai/SKILL.md) (`ai release update-draft --resources …` → `ai release deploy-draft`). See `references/examples/files.md` for the upload → attach → deploy sequence.

## Help

Every command supports `--help`:

```bash
cargo-ai content file upload --help
cargo-ai content library create --help
```

Files in this skill

  • SKILL.md6.3 KB
  • references/examples/files.md2.8 KB
  • references/response-shapes.md1.7 KB
  • references/troubleshooting.md1.3 KB
  • skill-metadata.json801 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.