Agent Pipeline
Development

Enforced multi-agent pipelines with mechanical guardrails for AI-assisted software development. Distributed as a Claude Code plugin.

click to copy /plugin install claude-apd@zstevovich-plugins
Interactive Demo Pipeline Runs Getting Started GitHub
The Pipeline
Every step mechanically enforced. No bypass from within Claude Code.
S
Spec
→
P
Plan
→
B
Builder
→
R
Reviewer
→
A
Adversarial
→
V
Verifier
→
✓
Commit
Mechanical Enforcement
Every rule backed by a hook script that blocks violations.
■

Pipeline Gate

Commits blocked without all 4 pipeline steps. Spec hash frozen mid-pipeline. Max 7 acceptance criteria.

⚠

Scope Guards

Agents scoped to specific files. Orchestrator blocked from writing code. Bash writes to pipeline state blocked.

✓

Spec Traceability

R* acceptance criteria traced to test files via @trace markers. Verifier blocks commit if any criterion untested.

★

Adversarial Review

Context-free code review with zero spec knowledge. Finds what code-reviewer misses. Opt-out per task via spec-card.

⚙

HMAC-Signed State

Pipeline .done files signed with compiled Go binary. Orchestrator cannot forge completion. Tamper-proof audit trail.

☰

Pipeline Report

Full recap dashboard in CLI. Step timing, spec coverage bars, adversarial insights, trend analysis, session stats.

Proven in Production
Real-world results across .NET and PHP projects
50+
pipeline runs tracked
22
guard scripts
8
skills
100%
success rate (last 20)
Get Started in 60 Seconds
# install the plugin
/plugin marketplace add zstevovich/claude-apd
/plugin install claude-apd@zstevovich-plugins

# configure your project
/apd-setup

# verify
bash .claude/bin/apd verify
Full Walkthrough
GitHub · Demo · Pipeline Runs · Changelog · Zoran Stevovic · MIT