77655c298c
Phase 8 Production Hardening with complete governance infrastructure: - Vault integration with tiered policies (T0-T4) - DragonflyDB state management - SQLite audit ledger - Pipeline DSL and templates - Promotion/revocation engine - Checkpoint system for session persistence - Health manager and circuit breaker for fault tolerance - GitHub/Slack integrations - Architectural test pipeline with bug watcher, suggestion engine, council review - Multi-agent chaos testing framework Test Results: - Governance tests: 68/68 passing - E2E workflow: 16/16 passing - Phase 2 Vault: 14/14 passing - Integration tests: 27/27 passing Coverage: 57.6% average across 12 phases Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
85 lines
3.1 KiB
Bash
Executable File
85 lines
3.1 KiB
Bash
Executable File
#!/bin/bash
|
|
#
|
|
# Context Checkpoint Skill for Claude Code
|
|
# =========================================
|
|
# Preserves session context and orchestrates sub-agent calls.
|
|
#
|
|
# Skill documentation: /opt/agent-governance/checkpoint/CLAUDE.md
|
|
#
|
|
# Commands:
|
|
# now [--notes "..."] Create checkpoint capturing current state
|
|
# load [checkpoint_id] Load latest or specific checkpoint
|
|
# diff [--from ID] [--to ID] Compare checkpoints
|
|
# list [--limit N] List available checkpoints
|
|
# summary --level <level> Context summary (minimal/compact/standard/full)
|
|
# report [--phase <phase>] Combined checkpoint + directory status report
|
|
# timeline [--limit N] Show checkpoint history with status changes
|
|
# auto-orchestrate --model <m> Delegate to OpenRouter models
|
|
# queue list|add|clear|pop Manage instruction queue
|
|
# prune [--keep N] Remove old checkpoints
|
|
#
|
|
# Examples:
|
|
# checkpoint now --notes "Before deploy"
|
|
# checkpoint load --json
|
|
# checkpoint summary --level compact
|
|
# checkpoint auto-orchestrate --model minimax --instruction "run tests"
|
|
#
|
|
|
|
set -euo pipefail
|
|
|
|
CHECKPOINT_DIR="/opt/agent-governance/checkpoint"
|
|
CHECKPOINT_SCRIPT="${CHECKPOINT_DIR}/checkpoint.py"
|
|
|
|
# Show help if no args or --help
|
|
if [[ $# -eq 0 ]] || [[ "${1:-}" == "--help" ]] || [[ "${1:-}" == "-h" ]]; then
|
|
cat << 'EOF'
|
|
Context Checkpoint Skill
|
|
========================
|
|
|
|
Commands:
|
|
now [--notes "..."] Create checkpoint (includes directory statuses)
|
|
load [checkpoint_id] [--json] Load checkpoint
|
|
diff [--from ID] [--to ID] Compare checkpoints
|
|
list [--limit N] [--json] List checkpoints
|
|
summary --level <level> Context summary
|
|
report [--phase <phase>] Combined checkpoint + directory status report
|
|
timeline [--limit N] Checkpoint history with status changes
|
|
auto-orchestrate --model <m> Delegate to AI models
|
|
queue list|add|clear|pop Manage instruction queue
|
|
prune [--keep N] Remove old checkpoints
|
|
|
|
Summary Levels:
|
|
minimal (~500 tokens) Phase + agent only
|
|
compact (~1000 tokens) + tasks + key variables
|
|
standard (~2000 tokens) + all tasks + dependencies
|
|
full (~4000 tokens) Complete context
|
|
|
|
Models for auto-orchestrate:
|
|
minimax Minimax-01 (default, 100K context)
|
|
gemini Gemini 2.0 Flash Thinking (free)
|
|
gemini-pro Gemini 2.5 Pro (highest capability)
|
|
|
|
Examples:
|
|
checkpoint now --notes "Phase 5 complete"
|
|
checkpoint load --json | jq .phase
|
|
checkpoint summary --level compact
|
|
checkpoint report # Combined status view
|
|
checkpoint report --phase in_progress # Filter by phase
|
|
checkpoint timeline --limit 5 # Recent history
|
|
checkpoint auto-orchestrate --model minimax -i "check status"
|
|
checkpoint queue add --instruction "run tests" --priority 5
|
|
|
|
Documentation: /opt/agent-governance/checkpoint/CLAUDE.md
|
|
EOF
|
|
exit 0
|
|
fi
|
|
|
|
# Verify checkpoint script exists
|
|
if [[ ! -f "${CHECKPOINT_SCRIPT}" ]]; then
|
|
echo "Error: checkpoint.py not found at ${CHECKPOINT_SCRIPT}"
|
|
exit 1
|
|
fi
|
|
|
|
# Execute checkpoint command
|
|
exec python3 "${CHECKPOINT_SCRIPT}" "$@"
|