auto
Smart router - Single entry point with natural language intent detection
GitHub-aware work monitor - triages issues, PRs, and CI failures
> /plugin marketplace add nyldn/claude-octopusHow it fires
How this command gets triggered: by you, by Claude, or both.
/sentinelContext preview
What this command does when you run it.
GitHub-aware work monitor - triages issues, PRs, and CI failures
command: sentinel disable-model-invocation: true description: GitHub-aware work monitor - triages issues, PRs, and CI failures version: 1.0.0 category: monitoring tags: [sentinel, github, triage, issues, ci, monitoring] created: 2026-02-21 updated: 2026-02-21
**Your first output line MUST be:** `๐ Octopus Sentinel`
GitHub-aware work monitor that triages issues, PRs, and CI failures. Sentinel observes and recommends workflows but never auto-executes them.
**When the user explicitly invokes `/octo:sentinel`, you MUST run the Sentinel orchestrator path below.** You are PROHIBITED from manually guessing repository status, skipping GitHub checks, or starting remediation without explicit user approval.
/octo:sentinel # One-time triage scan /octo:sentinel --watch # Continuous monitoring /octo:sentinel --canary # Post-deploy canary monitoring
For recurring triage, schedule Sentinel as a read-only Claude Code web or hosted task. Use `/octo:sentinel` for the normal scan and `/octo:sentinel --canary https://example.com` for post-deploy monitoring.
Scheduled Sentinel should stay triage-only. It may recommend `/octo:debug`, `/octo:review`, or `/octo:embrace`, but it must not start remediation unless the user explicitly asks for it.
| Source | Filter | Recommended Action | |--------|--------|--------------------| | Issues | `octopus` label | Classified via task type โ workflow recommendation | | PRs | Review requested | `/octo:review` for code review | | CI Runs | Failed status | `/octo:debug` for investigation |
| Variable | Default | Description | |----------|---------|-------------| | `OCTOPUS_SENTINEL_ENABLED` | `false` | Must be `true` to activate | | `OCTOPUS_SENTINEL_INTERVAL` | `600` | Poll interval for --watch mode (seconds) |
Sentinel is **triage-only**. It:
When the user invokes `/octo:sentinel`, you MUST:
OCTO_ROOT="${OCTO_ROOT:-${CLAUDE_PLUGIN_ROOT:-${HOME}/.claude-octopus/plugin}}"
OCTOPUS_SENTINEL_ENABLED=true bash "${OCTO_ROOT}/scripts/orchestrate.sh" sentinel $ARGUMENTSAfter triage, run the reaction engine to auto-respond to detected events:
# Check all active agents and fire reactions
OCTO_ROOT="${OCTO_ROOT:-${CLAUDE_PLUGIN_ROOT:-${HOME}/.claude-octopus/plugin}}"
REACTIONS="${OCTO_ROOT}/scripts/reactions.sh"
if [[ -x "$REACTIONS" ]]; then
"$REACTIONS" check-all
fiThis automatically forwards CI failure logs to agents, forwards review comments, and escalates stuck agents โ without requiring any new user commands.
Every AI model has blind spots. Claude Octopus supports twelve external provider integrations โ Codex, Antigravity CLI, Copilot, Qwen, Ollama, Perplexity, OpenRouter, OrcaRouter, OpenCode, Cursor CLI, Grok, and Kimi Code โ alongside the built-in Claude Code
Repo: nyldn/claude-octopus
Smart router - Single entry point with natural language intent detection
Start a creative thought partner brainstorming session
Switch Octopus model routing to the configured budget tier
[advanced] Activate destructive command warnings for the session
[advanced] Compatibility view of the deterministic Octopus usage report
Multi-LLM council for advice, decision support, implementation plans, and gated implementation