markhazlett

Dim Investigator — Development agent for Claude Code

Development community

Read-only stage-2 reviewer for /deep-review LOW/MED false-positive dimensions — types, observability, tests, api-drift, deps, a11y, dead-code, docs.

How to install Dim Investigator

Installs to ~/.claude/agents/markhazlett-agent-harness-dim-investigator.md

Terminal
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/markhazlett/agent-harness/HEAD/agents/dim-investigator.md -o ~/.claude/agents/markhazlett-agent-harness-dim-investigator.md

Restart Claude Code, or start a new session, for it to be picked up.

What Dim Investigator does

name: dim-investigator description: Read-only stage-2 reviewer for /deep-review LOW/MED false-positive dimensions — types, observability, tests, api-drift, deps, a11y, dead-code, docs. model: sonnet disallowedTools:

  • Edit
  • Write
  • MultiEdit
  • NotebookEdit

Dimension Investigator

You are a read-only review subagent dispatched by `/deep-review` (stage 2) for one of the binary-verdict dimensions: types, observability, tests, api-drift, deps, a11y, dead-code, or docs. These dime

Alternatives in Development

  • Docs Impact Agent — Finds repository documentation made false by a change and missing documentation required to use or operate a n 2.2k ★
  • Workflow Analyzer — Evaluates technical topics, proposals, or decisions across multiple dimensions with evidence-based scoring and 530 ★
  • A11y Expert 470 ★

Full documentation available on GitHub

View Source Repository

Related Agents

Dim Investigator Deep — Development agent for Claude Code

Dim Investigator Deep

Read-only stage-2 reviewer for /deep-review HIGH false-positive deep-thinking dimensions — structural, perform

Development community
Thrunt False Positive Auditor — Development agent for Claude Code

Thrunt False Positive Auditor

Fills Nyquist validation gaps by generating tests and verifying coverage for phase requirements

Development community
Bench Reviewer — Development agent for Claude Code

Bench Reviewer

Reviews eval suite quality — fixture/rubric consistency, scoring accuracy, false positive traps, difficulty ca

Development community
Dead Code Hunter — Development agent for Claude Code

Dead Code Hunter

Verifies dead code findings from knip/grep. Reads actual source files to filter false positives. Cites file:li

Development community
Markdown Investigator — Development agent for Claude Code

Markdown Investigator

Autonomous analyzer that determines if markdown errors are fixable or false positives

Development community
Genesis Investigator — Development agent for Claude Code

Genesis Investigator

Diagnoses Genesis subsystem failures. Use when something is broken, degraded, or reporting unexpected state. K

Development community

Related Skills

All Recommendations — Development skill for Claude Code

All Recommendations

Launch a team of implementation agents to apply ALL recommended changes from the most recent review — every se

Development community
Code Cleanup — Development skill for Claude Code

Code Cleanup

Deep codebase cleanup via eight parallel subagents that inherit this session's context. Each owns one concern

Development community
False Positive Triage — Data skill for Claude Code

False Positive Triage

Use when investigating a suspected false positive in drift detection, reducing false positive rates, or adding

Data community