// linkedin + x · one CLI for agents

LinkedIn and X, as one API for your agent.

One install. One login. Your agent reads, searches, and graphs LinkedIn and X in under sixty seconds — from any shell, any runtime.

no keys to manage · one bill · auditable

~/agent
added social-agent in 2.4s
✓ installed · binary: social
→ open https://social.dev/device?code=K3R-7QX
✓ signed in as cyrus@example.com · scope: read
[3 results · 412ms]
1. Shipping a CLI for agents that gives them...
2. Why every agent stack needs a graph layer...
3. Notes on hiring infra eng in the AI era...
 

// capabilities

Everything an agent needs to operate on the social graph.

A small, careful surface. Same flags across platforms. JSON in, JSON out.

// search

Search both networks.

One query shape across LinkedIn posts and X recent. Pipe straight into your agent's planner.

social linkedin search posts "AI infra"

// graph

Walk the graph.

Profiles, connections, employees. Enough relational signal to reason about who matters and why.

social linkedin users connections cyrus

// react

Read the room.

Pull a post's reactions and comments before you respond. Context, not vibes.

social linkedin posts reactions <urn>

// cross

X, in the same shape.

Home timeline, bookmarks, user tweets, search — the same flags and JSON output as LinkedIn.

social x timelines home --limit 20

// example · 23 lines

A real agent task, in shell you can read.

warm-contacts.sh

# Find hiring posts, fan out to reaction graphs,# emit a ranked list of warm contacts.social linkedin search posts "AI infra hiring" --limit 10 --json \  | jq -r '.posts[].urn' \  | xargs -I{} social linkedin posts reactions {} --json \  | jq -s '      [.[].reactions[] | { actor: .actor, type: .type }]        | group_by(.actor.id)        | map({ actor: .[0].actor, signals: length })        | sort_by(-.signals)        | .[0:15]    '
  • Composable with the tools you already use.

    Every command speaks JSON. Pipe through jq, fzf, your agent's planner — anything that reads stdin.

  • Same flags on LinkedIn and X.

    --limit, --account, --json, --no-cache. Learn the surface once; reuse it across networks.

  • Designed for agents first.

    Deterministic exit codes, structured errors, an MCP-shaped schema you can hand to a model.

// account safety

Built so your accounts outlive your prototype.

The unspoken question: will it kill my LinkedIn? Here's the short answer.

// rate

Rate-limited by default.

Conservative per-account budgets keep agents inside platform tolerances.

// scope

Scope-aware sessions.

Read scope is the default. Write requires an explicit upgrade you can revoke.

// oauth

Official OAuth where available.

Hosted auth for LinkedIn and X. No password handling, no scraped cookies.

// ship today

Ship your agent's LinkedIn integration today.

Two commands. No SDK to learn, no keys to rotate, no scraping to maintain.

install

bun install -g social-agentsocial login