spectra-tools

Appearance

Guide: Agent Integration with xapi-cli ​

xapi-cli is designed for machine-friendly use in agent pipelines.

1) Discover command capabilities ​

Use the built-in manifest to inspect commands:

bash
# Compact command index
xapi-cli --llms

# Full manifest with args, env vars, and output fields
xapi-cli --llms-full

For command-level schema introspection:

bash
xapi-cli posts search --schema
xapi-cli users get --schema
xapi-cli dm send --schema

2) Execute with structured output ​

Use JSON output for deterministic parsing:

bash
xapi-cli posts search "ai agents" --max-results 10 --json
xapi-cli users get jack --json

For richer orchestration metadata, add --verbose.

3) Pick an auth profile for your agent ​

Read-only agent profile ​

Use bearer auth when the agent only needs read operations.

bash
export X_BEARER_TOKEN="your-app-bearer-token"

Read + write agent profile ​

Use OAuth user context when the agent may post or send DMs.

bash
export X_ACCESS_TOKEN="your-oauth2-user-access-token"

Write commands (posts create, posts delete, dm send) require X_ACCESS_TOKEN.

4) Example agent loop ​

A minimal monitor-and-respond pattern:

bash
# discover
xapi-cli --llms-full

# gather context
xapi-cli users posts abstractchain --max-results 5 --json
xapi-cli posts search "abstract chain" --max-results 10 --json

# optional action (requires X_ACCESS_TOKEN)
xapi-cli posts create --text "Tracking updates from the ecosystem." --json

5) Reliability tips ​

  • Parse command output as JSON, not terminal text.
  • Validate command arguments using --schema before execution.
  • Separate read-only and write-capable credentials in your runtime.
  • Handle endpoint-specific auth limitations (for example, users search may require X_ACCESS_TOKEN).