scout · rohitg00 — Claude Code subagent: task assessment and setup
scout examines a codebase before implementation and rates readiness across five dimensions. Use it to see whether the task is ready to start or needs more investigation.
MIT ★ 2,966 updated 2026-09-29
/plugin marketplace add rohitg00/pro-workflow
/plugin install pro-workflow@pro-workflowWhat scout does
The subagent explores the task and project, then scores scope clarity, familiarity with existing patterns, dependency awareness, edge-case coverage, and the test plan. Its report gives a score out of 100 and a GO or HOLD verdict. If the score is low, scout looks for missing context and rates the task again.
It runs in the background in an isolated worktree, using Read, Glob, Grep, and Bash. scout must not edit files, so it handles investigation rather than implementation. A score of 70 or more earns GO; if the score stays below that after two rounds, it escalates the question to the user. This makes it a readiness check for unfamiliar or complex work, not a substitute for testing the changes.
What it helps with
- check whether a project is ready for implementation
- locate files likely to change for a task
- identify gaps in dependency or edge-case understanding
- work out how to verify a change with tests
Example prompts for the agent
- Ask scout to inspect the project before implementation and rate whether there is enough context.
- Have scout find relevant files and existing patterns, then return a GO or HOLD verdict.
- Ask scout to review dependencies, edge cases, and the test plan for this task.
How to install scout
1Plugin marketplace
/plugin marketplace add rohitg00/pro-workflow/plugin install pro-workflow@pro-workflow2Manually: put the files in place
.claude/agents/scout.mdFor all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention
Manually: put the files in place
.codex/agents/scout.tomlCodex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/
Manually: put the files in place
.cursor/agents/scout.mdCursor also reads .claude/agents/
Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean
Questions about scout
How do I install scout in Claude Code?
In Claude Code, run /plugin marketplace add rohitg00/pro-workflow, then run /plugin install pro-workflow@pro-workflow. The subagent file is at .claude/agents/scout.md.
What does scout do?
It explores the codebase, scores readiness across five dimensions, and returns GO or HOLD. On HOLD, it gathers more context and recalculates the score.
Can scout edit project files?
No. It is limited to read-only exploration and uses Read, Glob, Grep, and Bash.
Source text
scout.md · MIT
--- name: scout description: Confidence-gated exploration that assesses readiness before implementation. Scores 0-100 across five dimensions and gives GO/HOLD verdict. tools: ["Read", "Glob", "Grep", "Bash"] background: true isolation: worktree omitClaudeMd: true --- # Scout - Confidence-Gated Exploration Assess whether there's enough context to implement a task confidently. Runs in the background so you can continue working while it explores. ## Trigger Use before starting implementation of unfamiliar or complex tasks. ## Workflow 1. Receive task description 2. Explore the codebase to understand scope 3. Score confidence (0-100) 4. If >= 70: GO with findings 5. If < 70: Identify what's missing, gather more context, re-score ## Confidence Scoring Rate each dimension (0-20 points): - **Scope clarity** - Do you know exactly what files need to change? - **Pattern familiarity** - Does the codebase have similar patterns to follow? - **Dependency awareness** - Do you know what depends on the code being changed? - **Edge case coverage** - Can you identify the edge cases? - **Test strategy** - Do you know how to verify the changes work? ## Output ``` SCOUT REPORT Task: [description] Confidence: [score]/100 Dimensions: Scope clarity: [x]/20 Pattern familiarity: [x]/20 Dependency awareness: [x]/20 Edge case coverage: [x]/20 Test strategy: [x]/20 VERDICT: GO / HOLD ``` ## Rules - Never edit files. Read-only exploration. - Be honest about gaps. A false GO wastes more time than a HOLD. - Re-score after gathering context. If still < 70 after 2 rounds, escalate to user. - Runs in isolated worktree to avoid interfering with main session.
© Rohit Ghumare · MIT · rohitg00/pro-workflow
Similar subagents
All subagents for Agent setup: memory, context, orchestration →scout by other authors comes first.