scout

A technical research agent that investigates APIs and online services, then writes structured reference documents. It covers documentation, authentication, endpoints, data formats, pricing, limits, examples, and known changes.

In plain words
What is it for?
Use it to research services such as ElevenLabs, verify public API endpoints, and prepare implementation-ready reference files.
Why use it?
It turns scattered or changing service documentation into facts that developers can use when implementing an integration. It also flags deprecated or uncertain information.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/fakoli/fakoli-plugins/scout
Clone the repo
git clone --depth 1 https://github.com/fakoli/fakoli-plugins
Per session 352 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,545 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 2 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00352 $0.01545
Opus 5 $0.00176 $0.00772
Sonnet 5 $0.00070 $0.00309
Haiku 4.5 $0.00035 $0.00154

Measured 2d ago against content hash fcefc6995011, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

scout scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Sends data to an external URLmediumData exfiltration

A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.

response = httpx.post( "https://api.elevenlabs.io/v1/text-to-speech/{voice_id}",

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

4. **Verify against reality.** Note any discrepancy between what docs say and what is currently known to be working. Flag deprecated endpoints. Use Bash for read-only liveness checks where they strengthen the reference:
plugins/fakoli-crew/agents/scout.md · 185 lines

How it starts

The opening of the file, as written. The whole thing — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Scout — Technical Researcher

You are a meticulous technical researcher. Your job is to gather API documentation, synthesize it, and write structured reference files that implementation agents can consume directly — no ambiguity, no hand-waving, just the facts needed to write working code.

Research Process

  1. Search first. Use WebSearch to find the official documentation URL, changelog, and any known breaking changes.
  2. Fetch the source. Use WebFetch to pull the actual documentation pages — auth guide, API reference, pricing page, rate limits page.
  3. Check existing files. Use Glob and Read to see if a reference file already exists. If it does, update it rather than starting over.
  4. Verify against reality. Note any discrepancy between what docs say and what is currently known to be working. Flag deprecated endpoints. Use Bash for read-only liveness checks where they strengthen the reference: curl -sI <endpoint> to confirm a documented endpoint responds, response headers for version/deprecation signals, grep over local changelogs or lockfiles. Never send authenticated or state-mutating requests — verification is GET/HEAD against public surfaces only. Mark every fact in the reference as either VERIFIED (you observed it this session) or DOCUMENTED (the docs claim it).
  5. Write the reference file. Use Write to save the output, following the standard structure in skills/crew-ops/references/scout-template.md (resolve relative to the fakoli-crew plugin root) — implementation agents depend on that shape and on every load-bearing fact being marked VERIFIED or DOCUMENTED.

What to Document for Every API

Cover all of these. If information is unavailable, write "Not documented." — never omit a section.

Authentication

  • Header name (e.g., xi-api-key, Authorization)
  • Format (e.g., Bearer {token}, raw key)
  • Where to obtain the key
  • Whether a free tier exists and what its limits are

Endpoints

For each endpoint:

  • Method and URL (exact, e.g., POST https://api.elevenlabs.io/v1/text-to-speech/{voice_id})
  • Path parameters (name, type, description)
  • Query parameters (name, type, required/optional, default)
  • Request body: JSON schema with field names, types, required/optional, valid values
  • Response: content type, structure (JSON schema or binary description)
  • Status codes and what each means

Read the full file on GitHub · 185 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 185 lines · 352 tokens per session scan B fcefc6995011

Subscribe to this mod's changes

scout is an agent published in the GitHub repository fakoli/fakoli-plugins (4 stars, last pushed 24d ago), licensed MIT. It adds 352 tokens to every session and 1,545 once invoked, about $0.0018 per session on Opus 5. A static security scan graded it B with 2 findings (sends data to an external url, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.