cli.md

hu CLI Reference

Complete command reference for the hu CLI.

Installation

cargo install hu

Or from source:

cargo install --path .

Configuration

Credentials: ~/.config/hu/credentials.toml Settings: ~/.config/hu/settings.toml

---

Jira

Manage Jira tickets without leaving the terminal. OAuth 2.0 authentication, sprint views, JQL search, and ticket updates.

hu jira auth                   # OAuth 2.0 authentication
hu jira tickets                # List my tickets in current sprint
hu jira sprint                 # Show all issues in current sprint
hu jira search <query>         # Search tickets using JQL
hu jira show <ticket>          # Show ticket details
hu jira update <ticket>        # Update a ticket
  --summary <text>             #   New summary/title
  --status <status>            #   New status (transition)
  --assign <user>              #   Assign to user (or "me")

GitHub

GitHub workflow integration. List PRs, monitor CI runs, extract test failures, and get AI-ready investigation context.

hu gh login -t <PAT>           # Authenticate with PAT
hu gh prs                      # List your open PRs
hu gh runs [ticket]            # List workflow runs
  -s, --status <status>        #   Filter: queued, in_progress, completed, success, failure
  -b, --branch <name>          #   Filter by branch
  -r, --repo <owner/repo>      #   Repository
  -n, --limit <n>              #   Max results (default: 20)
  -j, --json                   #   Output as JSON
hu gh failures                 # Extract test failures from CI
  --pr <number>                #   PR number (default: current branch)
  -r, --repo <owner/repo>      #   Repository
hu gh fix                      # Analyze CI failures, output investigation context
  --pr <number>                #   PR number
  --run <id>                   #   Workflow run ID
  -b, --branch <name>          #   Branch name
  -r, --repo <owner/repo>      #   Repository
  -j, --json                   #   Output as JSON

Slack

Slack workspace access. Send messages, search history, list channels, and bulk-tidy unread channels.

hu slack auth                  # Authenticate with Slack
  --token <xoxb-...>           #   Bot token
  --user-token <xoxp-...>      #   User token (for search)
hu slack channels              # List channels
hu slack info <channel>        # Show channel details
hu slack send <channel> <msg>  # Send message
hu slack history <channel>     # Show message history
  --limit <n>                  #   Number of messages (default: 20)
hu slack search <query>        # Search messages
  -n, --count <n>              #   Max results (default: 20)
hu slack users                 # List users
hu slack config                # Show configuration status
hu slack whoami                # Show current user info
hu slack tidy                  # Mark channels as read if no mentions
  --dry-run                    #   Preview without marking

PagerDuty

On-call schedules, active alerts, and incident management.

hu pagerduty auth <token>      # Set API token
hu pagerduty config            # Show configuration status
hu pagerduty oncall            # Show who's currently on call
  -p, --policy <id>            #   Filter by escalation policy
  -s, --schedule <id>          #   Filter by schedule
  --json                       #   Output as JSON
hu pagerduty alerts            # List active alerts
  -l, --limit <n>              #   Max alerts (default: 25)
  --json                       #   Output as JSON
hu pagerduty incidents         # List incidents with filters
  -s, --status <status>        #   Filter: triggered, acknowledged, resolved, active
  -l, --limit <n>              #   Max incidents (default: 25)
  --json                       #   Output as JSON
hu pagerduty show <id>         # Show incident details
  --json                       #   Output as JSON
hu pagerduty whoami            # Show current user info
  --json                       #   Output as JSON
hu pd ...                      # Alias: pd -> pagerduty

Sentry

Error tracking integration. List unresolved issues, view error details, and browse event history.

hu sentry auth <token>         # Set auth token
hu sentry config               # Show configuration status
hu sentry issues               # List unresolved issues
hu sentry show <id>            # Show issue details
hu sentry events <id>          # List events for an issue

NewRelic

Application performance monitoring. Query NRQL, list incidents and issues.

hu newrelic auth <key>         # Set API key
  --account <id>               #   Account ID (required)
hu newrelic config             # Show configuration status
hu newrelic issues             # List recent issues
  --limit <n>                  #   Max issues (default: 25)
hu newrelic incidents          # List recent incidents
  --limit <n>                  #   Max incidents (default: 25)
hu newrelic query <nrql>       # Run NRQL query
hu nr ...                      # Alias: nr -> newrelic

Pipeline (AWS CodePipeline)

Monitor AWS CodePipeline deployments.

hu pipeline list               # List all pipelines
  -r, --region <region>        #   AWS region
  --json                       #   Output as JSON
hu pipeline status <name>      # Show pipeline status (stages and actions)
  -r, --region <region>        #   AWS region
  --json                       #   Output as JSON
hu pipeline history <name>     # Show execution history
  -r, --region <region>        #   AWS region
  -l, --limit <n>              #   Max results (default: 10)
  --json                       #   Output as JSON

EKS

Kubernetes pod access for EKS clusters.

hu eks list                    # List pods in the cluster
  -n, --namespace <ns>         #   Namespace
  -A, --all-namespaces         #   All namespaces
  -c, --context <ctx>          #   Kubeconfig context
  --json                       #   Output as JSON
hu eks exec <pod>              # Execute command in pod (shell by default)
  -n, --namespace <ns>         #   Namespace
  -c, --container <name>       #   Container name
  --context <ctx>              #   Kubeconfig context
  -- <command>                 #   Command to run (default: /bin/sh)
hu eks logs <pod>              # Tail logs from a pod
  -n, --namespace <ns>         #   Namespace
  -c, --container <name>       #   Container name
  -f, --follow                 #   Follow log output
  --previous                   #   Previous container instance
  --tail <n>                   #   Lines from end
  --context <ctx>              #   Kubeconfig context

Data (Claude Code Sessions)

Sync and analyze Claude Code session data. Track usage, search history, monitor tool usage, analyze costs.

hu data sync                   # Sync Claude data to local database
  -f, --force                  #   Force full resync
  -q, --quiet                  #   Quiet output
hu data config                 # Show data configuration
  -j, --json                   #   Output as JSON
hu data session list           # List sessions
  -p, --project <dir>          #   Filter by project
  -n, --limit <n>              #   Max results (default: 20)
  -j, --json                   #   Output as JSON
hu data session read <id>      # Read session messages
  -j, --json                   #   Output as JSON
hu data session current        # Show current session
  -j, --json                   #   Output as JSON
hu data stats                  # Usage statistics
  -t, --today                  #   Today only
  -j, --json                   #   Output as JSON
hu data todos list             # List all todos
  -s, --status <status>        #   Filter by status
  -j, --json                   #   Output as JSON
hu data todos pending          # Show pending todos
  -p, --project <dir>          #   Filter by project
  -j, --json                   #   Output as JSON
hu data search <query>         # Search messages (full-text)
  -n, --limit <n>              #   Max results (default: 20)
  -j, --json                   #   Output as JSON
hu data tools                  # Tool usage statistics
  -t, --tool <name>            #   Detail for specific tool
  -j, --json                   #   Output as JSON
hu data errors                 # Extract errors from debug logs
  -r, --recent <days>          #   Days to look back (default: 7)
  -j, --json                   #   Output as JSON
hu data pricing                # Pricing analysis vs API costs
  -s, --subscription <tier>    #   Subscription tier (default: max20x)
  -b, --billing-day <day>      #   Billing day of month (default: 6)
  -j, --json                   #   Output as JSON
hu data branches               # Branch activity statistics
  -b, --branch <name>          #   Filter by branch
  -l, --limit <n>              #   Max results (default: 20)
  -j, --json                   #   Output as JSON

Utils

General-purpose utilities for web content, code search, and documentation.

# Fetch HTML and convert to markdown
hu utils fetch-html <url>
  -c, --content                # Extract main content only
  -s, --summary                # First N paragraphs + headings
  -l, --links                  # Extract links only
  -H, --headings               # Extract headings (outline)
  --selector <css>             # CSS selector (e.g., "article")
  -o, --output <file>          # Output to file
  -r, --raw                    # Raw output (no filtering)

# Smart grep with token-saving options
hu utils grep <pattern> [path]
  --refs                       # File:line references only
  --unique                     # Deduplicate similar matches
  --ranked                     # Sort by relevance
  -n, --limit <n>              # Limit results
  --signature                  # Function/class signature only
  -g, --glob <pattern>         # File glob (e.g., "*.rs")
  -i, --ignore-case            # Case insensitive
  --hidden                     # Include hidden files

# Web search (requires Brave Search API key)
hu utils web-search <query>
  -n, --results <n>            # Number of results (default: 3)
  -l, --list                   # Show results only (don't fetch)
  -o, --output <file>          # Output to file

# Documentation indexing
hu utils docs-index [path]     # Build heading index (JSON)
  -o, --output <file>          # Output index to file
hu utils docs-search <idx> <q> # Search docs index
  -n, --limit <n>              # Limit results
hu utils docs-section <f> <h>  # Extract section from markdown

Context Tracking

Track which files have been loaded in a Claude Code session to prevent duplicate reads.

hu context track <file...>     # Mark file(s) as loaded
hu context check <file...>     # Check if already in context
hu context summary             # Show all tracked files
hu context clear               # Reset tracking

Smart File Reading

Token-efficient file reading modes for AI agents.

hu read <file>
  -o, --outline                # Show functions, structs, classes
  -i, --interface              # Public API only
  -a, --around <line>          # Lines around line number
  -n, --context <n>            # Context lines (default: 10)
  -d, --diff                   # Git diff
  --commit <ref>               # Diff against commit (default: HEAD)

Install

Install hu hooks and slash commands to Claude Code.

hu install list                # List available components
hu install preview             # Show what would be installed
  -g, --global                 # Target ~/.claude (default)
  -l, --local                  # Target ./.claude
  --hooks-only                 # Install only hooks
  --commands-only              # Install only commands
hu install run                 # Install components
  -g, --global                 # Target ~/.claude (default)
  -l, --local                  # Target ./.claude
  -f, --force                  # Override existing files
  --hooks-only                 # Install only hooks
  --commands-only              # Install only commands

Development

just check    # fmt + clippy (must pass)
just test     # run tests (must pass)
just build    # build release