Nissth — AI skill for Claude Code
Deterministic execution framework for LLM coding agents.
How to install Nissth
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open uKmaz/Nissth and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Nissth does
Deterministic execution framework for LLM coding agents. Pre-structured context, append-only state, no exploration.
Alternatives in AI
- Prompt Tower — Context management for long-context LLMs, agents, and vibe coding 382 ★
- Cersei — The Rust SDK for building coding agents 274 ★
- Stacklit — One command gives AI agents instant codebase context 102 ★
README
Nissth
A deterministic execution framework for AI coding agents — operate, don't explore.
**What ships:** three Diagnostic Bridge stack bindings — [Spring Boot](Bindings/SpringBoot/), [Expo](Bindings/Expo/), [PostgreSQL](Bindings/Postgres/) — behind one `nissth-bridge` dispatcher, plus five repo checks under [`Tools/`](Tools/): project bootstrap and drift check, prose-vs-manifest validation, DBL frontmatter and freshness, phase-plan linting, and DBL regeneration worksheets.
**Verified:** Spring Boot 104/104 unit (7 further integration tests need Docker) · Expo 80/80 across 15 suites · PostgreSQL 107 pass / 18 skip (the skips need a live database). Measured from a fresh clone, which is what §8.x.6 requires before any phase may close.
**Not built yet:** hook-based enforcement under `.claude/`. Everything else described here exists and is tested.
This README is the 30-minute landing page. The complete reference is [`CLAUDE.md`](CLAUDE.md); the latest project state is the last entry of [`AgentReports/StatusUpdate.md`](AgentReports/StatusUpdate.md).
Table of contents
- What this is
- What this is not
- Reading the archived history
- Quick start — arriving agents
- Quick start — arriving humans
- The boot protocol
- The Loop
- Project structure
- Two layers above source — DBL and the Diagnostic Bridge
- The Implementation Template and plan-before-execute
- Reports — long-form companions to status entries
- Stack bindings — current state
- Installing and using a binding
- [Working in this repo as an agent](#working-
Related Skills
AI Workflow OS
Append-only ledger + 5-state machine that keeps 6 divisions honest across 5 AI platforms (Claude, ChatGPT, Gem
CTX Portable Project Context Engine Agent Harness
CTX is a lightweight, agent-agnostic project context engine and deterministic coding harness. It equips AI cod
Canny
Stops AI coding agents from claiming work is done without evidence. Deterministic hooks decide, TypeSafe's Jev
Topcode
topcode is a The next-generation CLI agent framework — model-agnostic, context-entropy-resistant, built on Sta
Hardstop Command
Hardstop is a pre-execution safety layer that blocks dangerous shell commands using pattern matching + LLM ana
Mem Compress
rtk-style pre-storage compression — collapses 3+ adjacent identical lines + merges adjacent key: value runs sh
Related Agents
Figma Discovery
Explore and map the current state of a Figma document. Use when the target has 8+ variants, unknown tree depth
Session Log.Agent
Use at the end of a work session to append a structured session log to ~/.local/state/session-logs/YYYY-MM-DD.
AI Verify
Evidence-first verification orchestrator. Dispatches specialist agents via the Agent tool: 1 deterministic age