Cc Sentinel — DevOps skill for Claude Code
Governance infrastructure for Claude Code — hooks, agents, and workflows that prevent the failure modes autonomous coding sessions actually hit.
How to install Cc Sentinel
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open turqoisehex/cc-sentinel and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Cc Sentinel does
Governance infrastructure for Claude Code — hooks, agents, and workflows that prevent the failure modes autonomous coding sessions actually hit.
Alternatives in DevOps
- Claude Code Router — Use Claude Code as the foundation for coding infrastructure, allowing you to decide how to interact with the m 30.1k ★
- Cccc — Coordinate your coding agents like a group chat — read receipts, delivery tracking, and remote ops from your p 1.1k ★
- Codeg — Aggregate and browse AI coding agent sessions (Claude Code, Codex, Gemini CLI, etc.) in one place 850 ★
README
cc-sentinel
[](https://github.com/turqoisehex/cc-sentinel/actions/workflows/ci.yml) [](LICENSE) [](https://github.com/turqoisehex/cc-sentinel)
**Governance infrastructure for Claude Code.** Hooks, agents, and workflows that prevent the failure modes autonomous coding sessions actually hit.
The Problem
Claude Code is powerful out of the box. But long, autonomous sessions surface real failure modes that no amount of prompting fixes:
| Failure Mode | What Happens | Without cc-sentinel | With cc-sentinel |
|---|---|---|---|
| Context loss | After compaction, Claude forgets what it was doing, repeats work, or contradicts earlier decisions | Hours of wasted compute, inconsistent output | Pre-compact hook saves state; post-compact hook restores it. Sessions survive compaction. |
| Work deferral | Claude writes "TODO: implement later" or "will add in next step" and never returns | Incomplete features shipped as "done" | Anti-deferral hook detects deferral language in every file write and warns immediately. |
| False completion | Claude claims a task is done without verifying | Bugs discovered in production, not development | Stop hook blocks completion claims without verification squad evidence on disk. Multi-model interleaved verification (Sonnet + Codex) catches bugs no single model finds alone. |
| Governance drift | Claude edits its own rules, CLAUDE.md, or config files mid-session | Guardrails silently disabled | File-protection hook blocks writes to protected files. Override requires explicit authorization marker. |
| Silent compaction | Context window fills with no warning; state is lost before it can be saved | Unrecoverable mid-task failure | Visual status bar + 5 graduated warnings at 50/ |
Related Skills
Symptomsync Review
Use when reviewing code, plans, or diffs in this repository. Apply it to frontend, SQL, AI-service, infrastruc
Brainvault
Personal memory layer for your Coding Agent — SQLite vault, MCP tools, and session hooks. Zero infrastructure.
Remote Vibe Station
Remote Vibe Station is a server-first AI coding workspace. Instead of running OpenCode on your laptop, you put
AI Contributing
Open-source multi-agent governance for running multiple AI coding agents in parallel on one GitHub repo: issue
Legalise
Open-source governance infrastructure for AI-assisted legal work
Runwall
The active firewall and execution governance gateway for AI agents. Run Cursor, Claude Code, Cline, and custom
Related Agents
Claude Code Hook Agent
Plays agent-specific sounds for the 6 hooks that actually fire in agent sessions
Calendar Agent
Builds a publishing calendar the creator can actually hit — cadence from the channel archetype, backwards-plan
Probe Canonical Edit
PLAN-020 Phase 0 item 5 governance probe — verifies that a native subagent attempting to Edit/Write a canonica