agentleFS
jackmazacjackmazac/exa-cli/skills/build-with-exa/SKILL.md

build-with-exa

Skill1 starsChanged 4 months agoMITView on GitHub

Use when building applications or agents that need Exa web search, known-URL extraction, or long-running research through the native stateless exa CLI.

  • Reads credentials

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. exa-labs/agent-skills★ 67same name
    2. MoizIbnYousaf/marketing-cli★ 33same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository2 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 4 are distinctive, and 7 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 7

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use exa as a JSON subprocess.

    line 10
  2. Distinctive:

    Only three Exa surfaces are available:

    line 12
  3. Distinctive:

    Default to Search; reserve Agent for multi-step research.

    line 20
  4. Distinctive:

    Never pass the API key on the command line, place it in JSON input, log the environment, or embed it in source code.

    line 31
  5. Distinctive:

    Use convenience arguments for common calls:

    line 35
  6. Very specific:

    Use --input <file|-> for advanced fields.

    line 43
  7. Very specific:

    Do not mix --input with positional arguments or command-specific convenience flags.

    line 56
  8. Very specific:

    Never repeat Agent creation after UPSTREAM_STATE_UNKNOWN.

    line 91
  9. Very specific:

    Use the exa-search skill for Search request fields, modes, filters, and result handling.

    line 95
  10. Very specific:

    Use the exa-contents skill for known-URL extraction, freshness, and per-URL statuses.

    line 96
  11. Very specific:

    Read common mistakes before constructing advanced --input objects.

    line 97

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Build with Exa CLI

Overview

Use exa as a JSON subprocess. Each invocation performs one bounded workflow, emits one envelope, and exits. It has no daemon, session, credential store, cache, or background process.

Only three Exa surfaces are available:

NeedCommand
Discover and rank web sourcesexa search
Extract content from known URLsexa fetch
Run multi-step research or enrichmentexa agent run

Default to Search; reserve Agent for multi-step research.

Prerequisites

Confirm installation and set the credential:

exa --version
export EXA_API_KEY="your_api_key_here"

Never pass the API key on the command line, place it in JSON input, log the environment, or embed it in source code.

Choose an Invocation Mode

Use convenience arguments for common calls:

exa search "latest grid-scale battery developments" --num-results 5
exa fetch "https://example.com/report" --max-characters 5000
exa agent run "Compare five storage technologies with evidence." --effort low

Use --input <file|-> for advanced fields. - reads one object from stdin:

exa search --input - <<'JSON'
{
  "query": "recent grid-scale battery deployments",
  "type": "deep",
  "numResults": 8,
  "contents": { "highlights": true }
}
JSON

Do not mix --input with positional arguments or command-specific convenience flags. Global flags such as --pretty, --verbose, and --timeout-ms may still be used.

Unknown input fields are forwarded. The CLI still owns transport: omit Search streaming and Agent stream.

Consume the Envelope

Stdout contains one newline-terminated document:

{
  "version": 1,
  "ok": true,
  "command": "search",
  "data": { "results": [] }
}

Failures replace data with error. Read fields under data, check the exit status and ok, and keep stderr separate. See CLI contract for defaults, errors, and exit codes.

Integrate with an Application

Spawn exa without a shell, send JSON through stdin, and parse stdout. Preserve EXA_API_KEY, but never serialize the environment.

See process integration for a complete TypeScript example.

Agent Lifecycle

exa agent run normally waits for a terminal result. If it returns a known running ID, continue with:

exa agent get agent_run_123
exa agent wait agent_run_123 --poll-interval-ms 4000
exa agent cancel agent_run_123

Never repeat Agent creation after UPSTREAM_STATE_UNKNOWN. See Agent workflows.

Surface-Specific Guidance

  • Use the exa-search skill for Search request fields, modes, filters, and result handling.
  • Use the exa-contents skill for known-URL extraction, freshness, and per-URL statuses.
  • Read common mistakes before constructing advanced --input objects.

Boundaries

The CLI does not expose Answer, Context, Monitors, Websets, compatibility endpoints, arbitrary endpoints, or Search streaming. Do not simulate them with undocumented flags.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.