一键导入
idumb-project-validation
PROJECT package for validating user projects - works in greenfield and brownfield, ensures governance without blocking development
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
PROJECT package for validating user projects - works in greenfield and brownfield, ensures governance without blocking development
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Validates iDumb plugin functionality through systematic pivotal trials and stress testing. Ensures all success criteria are met with concrete evidence. Use when testing any iDumb component or before considering implementation complete.
Guides the implementation of the iDumb plugin reboot following the 8-phase pivotal trial methodology. Use when building TypeScript engines, tools, and hooks for intelligent AI governance.
This skill should be used when the user asks to "orchestrate", "delegate", "coordinate agents", "enforce hierarchy", "mindful delegation", "hierarchical thinking", "chain of command", or "orchestrator patterns". Essential for coordinators and high-governance agents to maintain delegation integrity.
Complete iDumb governance protocols - hierarchical delegation, validation patterns, context anchoring, and expert-skeptic mode guidelines
Meta-skill for idumb-builder agent to transform confusing specifications into structured iDumb workflow modules with checkpoints, validation, and integration points. Use when creating reusable workflows, handling overlapping specs, or generating governance modules under .idumb/modules/. This skill enforces the iDumb vision: spec-driven development, context-first validation, drift detection, and hierarchical coordination.
This skill should be used when the user asks to "ingest meta-framework", "transform framework to idumb", "classify framework patterns", "create self-upgrading module", "meta-learning", "framework synthesis", "automated governance enhancement", "pattern extraction", "hierarchical framework analysis", or "iterative framework learning". Essential for idumb-builder to consume external frameworks like BMAD and transform them into iDumb-compatible governance patterns.
| name | idumb-project-validation |
| description | PROJECT package for validating user projects - works in greenfield and brownfield, ensures governance without blocking development |
| version | 1.0.0 |
| license | MIT |
| metadata | {"audience":"ai-agents","workflow":"project-validation","package":"PROJECT","activation":"coordinator-driven","modes":[{"pre-flight":"Before starting work"},{"continuous":"During development"},{"health-check":"On-demand status"}]} |
<project_detection>
greenfield_indicators:
strong:
- no_src_directory: true
- no_package_json: true
- empty_git_history: "< 5 commits"
- no_existing_governance: "no .planning/, no .idumb/"
moderate:
- recent_init: "created < 7 days ago"
- minimal_files: "< 20 files"
- no_tests: true
detection_command: |
FILES=$(find . -type f | grep -v node_modules | grep -v .git | wc -l)
COMMITS=$(git log --oneline 2>/dev/null | wc -l)
[ "$FILES" -lt 20 ] && [ "$COMMITS" -lt 5 ] && echo "GREENFIELD"
brownfield_indicators:
strong:
- existing_src: true
- package_json_with_deps: "> 5 dependencies"
- git_history: "> 20 commits"
- existing_tests: true
moderate:
- documentation_exists: "README.md with content"
- ci_cd_configured: ".github/workflows/ exists"
- multiple_contributors: "> 1 author in git"
detection_command: |
COMMITS=$(git log --oneline 2>/dev/null | wc -l)
DEPS=$(jq '.dependencies | length' package.json 2>/dev/null || echo 0)
[ "$COMMITS" -gt 20 ] || [ "$DEPS" -gt 5 ] && echo "BROWNFIELD"
complexity_levels:
simple:
files: "< 50"
languages: 1
frameworks: "0-1"
team_size: 1
moderate:
files: "50-500"
languages: "1-3"
frameworks: "1-3"
team_size: "1-5"
complex:
files: "> 500"
languages: "> 3"
frameworks: "> 3"
team_size: "> 5"
detection_command: |
FILES=$(find . -type f | grep -v node_modules | grep -v .git | wc -l)
LANGS=$(find . -type f -name "*.ts" -o -name "*.js" -o -name "*.py" -o -name "*.go" |
sed 's/.*\.//' | sort -u | wc -l)
if [ "$FILES" -lt 50 ] && [ "$LANGS" -le 1 ]; then
echo "SIMPLE"
elif [ "$FILES" -lt 500 ] && [ "$LANGS" -le 3 ]; then
echo "MODERATE"
else
echo "COMPLEX"
fi
</project_detection>
<validation_modes>
When: Before starting any work on a project
Purpose: Ensure environment is ready for iDumb governance
pre_flight_checks:
environment:
- opencode_available: "which opencode"
- node_version: ">= 18"
- git_available: "which git"
- git_initialized: "test -d .git"
idumb_state:
- idumb_initialized: "test -d .idumb"
- state_valid: "jq . .idumb/brain/state.json"
- config_valid: "jq . .idumb/brain/config.json"
project_state:
- no_uncommitted_changes: "git status --porcelain"
- on_feature_branch: "git branch --show-current"
- upstream_current: "git fetch && git status -uno"
governance_state:
- agents_available: "ls .opencode/agents/idumb-*.md"
- commands_available: "ls .opencode/commands/idumb/*.md"
- tools_available: "ls .opencode/tools/idumb-*.ts"
pre_flight_output:
pass:
message: "Environment ready for iDumb governance"
proceed: true
fail:
message: "Pre-flight checks failed"
issues: "list of failed checks"
fixes: "suggested fixes for each"
proceed: false
partial:
message: "Some checks failed but can proceed"
warnings: "list of warnings"
proceed: true
When: During active development, triggered by coordinator
Purpose: Ensure ongoing work doesn't violate governance
continuous_validation:
triggers:
micro_triggers:
- file_saved: "*.ts, *.md, *.json"
- command_executed: "/idumb:*"
- agent_spawned: true
batch_triggers:
- phase_transition: true
- milestone_complete: true
- commit_made: true
coordinator_decision:
use_micro_when:
- editing_governance_files: true
- high_risk_action: true
- previous_validation_failed: true
use_batch_when:
- phase_boundary: true
- multiple_changes: "> 5 files"
- time_elapsed: "> 30 min"
continuous_checks:
always_check:
- state_consistency: "state.json valid"
- no_conflicts: "no circular refs"
- chain_integrity: "delegation valid"
on_file_change:
- file_in_scope: "matches governance patterns"
- no_secrets: "scan for API keys"
- syntax_valid: "YAML/JSON parseable"
on_phase_transition:
- phase_complete: "all tasks done"
- artifacts_exist: "expected files created"
- verification_passed: "quality gates met"
When: On-demand status check
Purpose: Comprehensive project health assessment
health_check:
categories:
governance_health:
- idumb_state: "valid, fresh"
- agents_synced: "src/ matches .opencode/"
- config_current: "version matches"
project_health:
- tests_passing: "npm test"
- build_working: "npm run build"
- no_lint_errors: "npm run lint"
integration_health:
- git_clean: "no uncommitted changes"
- branch_current: "up to date with main"
- dependencies_fresh: "npm outdated"
health_score:
calculation: |
GOVERNANCE=$(check_governance_health) # 0-100
PROJECT=$(check_project_health) # 0-100
INTEGRATION=$(check_integration_health) # 0-100
SCORE=$((GOVERNANCE * 0.3 + PROJECT * 0.4 + INTEGRATION * 0.3))
ratings:
excellent: ">= 90"
good: "70-89"
fair: "50-69"
poor: "< 50"
</validation_modes>
<greenfield_bootstrap>
Set up optimal iDumb governance from scratch
greenfield_bootstrap:
step_1_detect:
action: "Confirm greenfield status"
command: "run project detection"
output: "GREENFIELD confirmed"
step_2_initialize:
action: "Initialize iDumb framework"
command: "/idumb:init"
creates:
- ".idumb/brain/state.json"
- ".idumb/brain/config.json"
- ".idumb/project-output/"
step_3_configure:
action: "Configure for project type"
prompts:
- "Project name?"
- "Primary language?"
- "Framework (if any)?"
- "Governance level (light/standard/strict)?"
output: "config.json updated"
step_4_scaffold:
action: "Create initial planning structure"
command: "/idumb:new-project"
creates:
- ".planning/PROJECT.md"
- ".planning/ROADMAP.md (if full governance)"
step_5_validate:
action: "Run initial validation"
command: "/idumb:validate"
expect: "all checks pass"
step_6_anchor:
action: "Create bootstrap anchor"
content: "Greenfield bootstrap complete"
priority: "high"
governance_levels:
light:
description: "Minimal governance for small/solo projects"
features:
- basic_state_tracking
- simple_validation
- manual_phase_transitions
agents_active: 5 # core only
standard:
description: "Balanced governance for typical projects"
features:
- full_state_tracking
- continuous_validation
- semi-auto_phase_transitions
agents_active: 12 # core + planning
strict:
description: "Full governance for complex/team projects"
features:
- comprehensive_tracking
- real_time_validation
- auto_phase_transitions
- skeptic_review
agents_active: 22 # all agents
</greenfield_bootstrap>
<brownfield_integration>
Integrate iDumb governance into existing project without disruption
brownfield_integration:
step_1_assess:
action: "Assess existing project"
checks:
- existing_governance: ".planning/, .gsd/, other"
- tech_stack: "languages, frameworks"
- test_coverage: "existing tests"
- ci_cd: "automation in place"
output: "assessment_report.json"
step_2_plan_integration:
action: "Create integration plan"
considers:
- existing_workflows: "preserve what works"
- migration_path: "incremental adoption"
- conflict_avoidance: "no breaking changes"
output: "integration_plan.md"
step_3_minimal_init:
action: "Initialize iDumb minimally"
command: "/idumb:init --brownfield"
creates:
- ".idumb/brain/state.json (minimal)"
- ".idumb/brain/config.json (adaptive)"
does_not_modify:
- existing_source_code
- existing_config_files
- existing_ci_cd
step_4_shadow_mode:
action: "Run in shadow mode first"
duration: "1-3 sessions"
behavior:
- observe: "log would-be validations"
- learn: "detect patterns"
- suggest: "offer improvements"
- block: "never"
output: "shadow_mode_report.md"
step_5_gradual_activation:
action: "Activate governance incrementally"
phases:
- phase_1: "state tracking only"
- phase_2: "add validation (non-blocking)"
- phase_3: "add auto-activation hooks"
- phase_4: "full governance (optional)"
step_6_validate_integration:
action: "Verify no disruption"
checks:
- tests_still_pass: "npm test"
- build_still_works: "npm run build"
- ci_cd_green: "no pipeline failures"
- team_workflow_preserved: "existing processes work"
conflict_avoidance:
existing_governance:
if_planning_exists:
action: "Integrate, don't replace"
strategy: "iDumb reads .planning/, adds .idumb/"
if_gsd_exists:
action: "Migrate if requested"
strategy: "Archive GSD, adopt iDumb"
if_custom_exists:
action: "Coexist"
strategy: "iDumb operates independently"
existing_tooling:
preserve:
- "ESLint config"
- "Prettier config"
- "CI/CD pipelines"
- "Git hooks (add, don't replace)"
existing_workflows:
respect:
- "Team branching strategy"
- "PR review process"
- "Release workflow"
</brownfield_integration>
<continuous_loop>
continuous_validation_loop:
trigger_sources:
- file_watcher: "src/, .planning/, .idumb/"
- command_hook: "after /idumb:* commands"
- phase_hook: "on phase transitions"
- time_based: "every 30 min if active"
loop_controller:
on_trigger:
- evaluate_trigger_type
- select_validation_mode
- run_appropriate_checks
- process_results
on_pass:
- log_success
- update_last_validation
- continue_development
on_fail:
- classify_failures
- attempt_self_heal
- report_remaining
- block_if_critical
on_stall:
- checkpoint_state
- report_to_user
- offer_options
non_blocking_validation:
principle: "Validation informs, rarely blocks"
blocking_conditions:
block_when:
- critical_security_issue: true
- governance_corruption: true
- circular_dependency: true
warn_when:
- high_severity_gap: true
- integration_below_threshold: true
- stale_context: true
inform_when:
- medium_severity_gap: true
- style_inconsistency: true
- optimization_opportunity: true
user_control:
override: "--force to bypass warnings"
disable: "--no-validate to skip"
strict: "--strict to block on warnings"
</continuous_loop>
<opencode_compatibility>
opencode_certification:
environment_compatibility:
- works_in_tui: "no console.log pollution"
- respects_opencode_config: "reads .opencode/ correctly"
- tools_export_correctly: "uses tool() wrapper"
- agents_spawn_correctly: "task delegation works"
cli_compatibility:
- commands_registered: "/idumb:* available"
- help_works: "/idumb:help shows all commands"
- arguments_parsed: "flags and args work"
plugin_compatibility:
- hooks_fire: "session start/end events"
- state_persists: "survives sessions"
- no_memory_leaks: "clean shutdown"
multi_project_compatibility:
- isolated_state: "per-project .idumb/"
- no_global_pollution: "doesn't affect other projects"
- portable: "works when project moved"
compatibility_checks:
check_opencode_version:
command: "opencode --version"
minimum: "0.1.0"
check_tool_loading:
command: "opencode tools list | grep idumb"
expect: "all idumb-* tools listed"
check_agent_loading:
command: "opencode agents list | grep idumb"
expect: "all idumb-* agents listed"
check_command_loading:
command: "opencode commands list | grep idumb"
expect: "all /idumb:* commands listed"
check_plugin_loading:
command: "opencode plugins list | grep idumb-core"
expect: "idumb-core.ts loaded"
</opencode_compatibility>
<integration_points>
framework_integration:
reads_from:
- ".idumb/brain/state.json"
- ".idumb/brain/config.json"
- ".opencode/agents/idumb-*.md"
- ".opencode/commands/idumb/*.md"
- ".opencode/tools/idumb-*.ts"
writes_to:
- ".idumb/brain/governance/validations/"
- ".idumb/brain/governance/health-checks/"
- ".idumb/brain/history/"
triggers:
- "/idumb:validate"
- "/idumb:status"
- "/idumb:health-check"
triggered_by:
- "Session start (auto)"
- "Phase transition (auto)"
- "User command (manual)"
project_integration:
respects:
- existing_tooling: "ESLint, Prettier, etc."
- existing_ci_cd: "GitHub Actions, etc."
- existing_governance: ".planning/, etc."
- existing_tests: "test runners"
enhances:
- adds_governance: ".idumb/ directory"
- adds_validation: "continuous checks"
- adds_guidance: "step-by-step workflows"
never:
- modifies_source_code: "without explicit request"
- breaks_existing_workflow: "preserves team process"
- conflicts_with_tooling: "coexists peacefully"
</integration_points>
# Pre-flight check before starting work
/idumb:pre-flight
# Run health check
/idumb:health-check
# Integration test for brownfield
/idumb:integration-test
# Bootstrap greenfield project
/idumb:init --greenfield
# Integrate with brownfield project
/idumb:init --brownfield
# In coordinator decisions
on_session_start:
- run: "pre-flight check"
- if_fail: "report issues"
on_phase_transition:
- run: "batch validation"
- if_fail: "block transition"
on_commit:
- run: "health check"
- if_warn: "log warnings"
<success_criteria>
Skill: idumb-project-validation v1.0.0 - PROJECT Package