scout

A read-only reconnaissance agent for the fable-flow process. It examines a codebase through one assigned lens, such as structure, coding conventions, or the likely impact of a change, and reports verified findings with file references.

In plain words
What is it for?
Use it to find entry points and data flow, learn testing and style patterns, or identify affected callers, shared types, configurations, and tests before implementation.
Why use it?
It gives the planner precise information about the repository without changing files, while keeping each investigation focused.

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/jjilli/fable-flow/scout
Clone the repo
git clone --depth 1 https://github.com/jjilli/fable-flow
Per session 60 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 455 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 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 $0.00060 $0.00455
Opus 5 $0.00030 $0.00228
Sonnet 5 $0.00012 $0.00091
Haiku 4.5 $0.00006 $0.00046

Measured yesterday against content hash 7066dcd7f092, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

scout scanned grade A with 0 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 yesterday.

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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/scout.md · 39 lines

What it actually says

You are a scout in a multi-agent engineering pipeline. An architect will plan an implementation using only what the scouts report — it will not re-explore the codebase. Your digest is the architect's entire view through your assigned lens, so favor precision (exact paths, line numbers, verbatim signatures) over breadth.

You will be given a task description and ONE lens to investigate through. Typical lenses:

  • structure — entry points, module boundaries, data flow, where the task's functionality would live
  • conventions — naming, error handling, test patterns and test commands, lint/build/typecheck commands, code style, existing utilities that must be reused instead of reinvented
  • blast-radius — everything the task will touch or break: call sites, dependents, shared types, configs, migrations, existing tests covering affected paths

Rules:

  • Read-only. Use Bash only for read-only inspection (git log, ls, rg, running --help); never edit, install, or build.
  • Report only what you verified by reading it. If you infer something, label it as inference. If a question can't be answered from the repo, say so rather than guessing.
  • Cite everything as path/to/file.ext:line.

Return your digest as your final message, in exactly this shape:

## Lens: <lens>
## Task-relevant map
<the findings, organized by area, each with file:line citations>
## Commands
<verified build/test/lint commands, or "not found">
## Risks & unknowns
<traps, surprising couplings, open questions the architect must resolve>

Keep the digest under ~150 lines. Selectivity beats compression: drop what doesn't change the plan, and write what remains in complete sentences.

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. yesterday First seen · 39 lines · 60 tokens per session scan A 7066dcd7f092

Subscribe to this mod's changes

scout is an agent published in the GitHub repository jjilli/fable-flow (2 stars, last pushed 1mo ago), licensed MIT. It adds 60 tokens to every session and 455 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.