scout

A read-only research workflow for investigating questions about a codebase and delivering a factual report. It can trace how code works and record the relevant file paths and line references.

In plain words
What is it for?
Use it to answer questions about system behavior, map authentication or other code paths, and list API endpoints with their middleware chains. It can optionally validate findings with another agent.
Why use it?
It reduces the time spent searching through unfamiliar code and helps keep conclusions tied to evidence. It does not change code, create pull requests, or make commits.

Skill for Claude CodeCodex

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 skills/justinjdev/fellowship/scout
Any agent
npx skills add justinjdev/fellowship --skill scout
Clone the repo
git clone --depth 1 https://github.com/justinjdev/fellowship

Made for: Claude Code, Codex.

Per session 46 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,183 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 $0.00046 $0.01183
Opus 5 $0.00023 $0.00592
Sonnet 5 $0.00009 $0.00237
Haiku 4.5 $0.00005 $0.00118

Measured 2d ago against content hash 61615bde1ae6, 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 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.

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.

plugin/skills/scout/SKILL.md · 121 lines

How it starts

The opening of the file, as written. The whole thing — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Scout — Research & Analysis

Overview

Investigates questions and analyzes codebases without producing code changes. Runs autonomously through Investigate → (Validate) → Deliver phases. When used as a fellowship teammate, the scout agent definition enforces read-only tool access. Also works standalone.

When to Use

  • Research questions about a codebase ("how does X work?")
  • Deep analysis ("what are all the entry points for auth?")
  • Data collection ("list all API endpoints and their middleware chains")

Phase Flow

Investigate ──→ (Validate) ──→ Deliver

Process

Investigate

Goal: Gather thorough, factual findings about the question.

Actions:

  1. Invoke /council to load task-relevant context
  2. Use Explore agents (Agent tool, subagent_type=Explore, passing model: "haiku" — or models.explore from fellowship config if set) to scan relevant code paths
  3. Read key files, trace call chains, understand behavior
  4. Document findings with specific file paths and line references

Investigate must produce:

  • Specific file paths and line ranges for every claim
  • Clear explanation of how things work (not just where they are)
  • Constraints, edge cases, and dependencies identified
  • Confidence level for each finding (High/Medium/Low)

High confidence = verified by reading actual code at specific lines. Medium confidence = inferred from patterns, naming, or partial evidence. Low confidence = assumption based on conventions or incomplete information.

If findings are incomplete, keep investigating. Don't move to validation or delivery with gaps.

Validate (conditional)

Goal: Adversarially verify findings using a fresh subagent with no context pollution.

When to validate:

  • Deep analysis involving multiple systems or complex interactions
  • Findings that will inform architectural decisions or code changes
  • Questions where being wrong would waste significant downstream effort
  • Any Medium or Low confidence finding that appears in a conclusion or recommendation of the report (if it's load-bearing, it gets validated)

Read the full file on GitHub · 121 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 · 121 lines · 46 tokens per session scan A 61615bde1ae6

Subscribe to this mod's changes

scout is a skill published in the GitHub repository justinjdev/fellowship (5 stars, last pushed 19d ago), licensed Apache-2.0. It adds 46 tokens to every session and 1,183 once invoked, about $0.0002 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.