/component-activate - Full Component Lifecycle Management

Complete component lifecycle: inventory → register → activate → classify → improve → index → commit.

Usage

# Full lifecycle for a single component
/component-activate path/to/component.md

# Full lifecycle for a directory
/component-activate agents/

# Specific stages only
/component-activate path/ --inventory          # List components
/component-activate path/ --register           # Add to index
/component-activate path/ --classify           # MoE classification
/component-activate path/ --improve            # Quality enhancement
/component-activate path/ --commit "message"   # Git commit

# Skip stages
/component-activate path/ --no-classify        # Skip classification
/component-activate path/ --no-commit          # Don't auto-commit

# Dry run
/component-activate path/ --dry-run

System Prompt

EXECUTION DIRECTIVE: When the user invokes /component-activate, execute the following 7-stage workflow:

Stage 1: Inventory

# Count and list components at path
python3 ~/.coditect/scripts/component-indexer.py --stats
python3 ~/.coditect/scripts/component-indexer.py --type agent --list
python3 ~/.coditect/scripts/component-indexer.py --type skill --list
python3 ~/.coditect/scripts/component-indexer.py --type command --list

Output: Component counts by type, list of unregistered components

Stage 2: Register

# Index components into SQLite database
python3 ~/.coditect/scripts/component-indexer.py -i  # Incremental
# OR
python3 ~/.coditect/scripts/component-indexer.py     # Full rebuild

Output: Components registered in framework registry

Stage 3: Activate

Activation means the component is ready for use:

Has valid frontmatter (title, version, status: active)
Is indexed in the component database
Has been classified by MoE

# Check activation status
python3 ~/.coditect/scripts/component-indexer.py --search "component-name" --verbose

Activation Checklist:

Frontmatter complete (title, version, type, status)
File in correct directory (agents/, skills/, commands/)
Indexed in component database
MoE classified (moe_confidence, moe_classified fields)

Stage 4: Classify (MoE)

# Run MoE classification on component(s)
/classify path/to/component.md

# Or batch classify directory
/classify agents/

Auto-adds fields:

type - Component type
component_type - Specific type
keywords - Auto-extracted keywords
tags - Classification tags
moe_confidence - Classification confidence (0.0-1.0)
moe_classified - Date of classification

Stage 5: Improve (Optional)

# Run quality improvement analysis
python3 ~/.coditect/scripts/skill-pattern-analyzer.py --component path/to/component.md

# Or use the improve skill
/agent component-enhancer "Improve component: path/to/component.md"

Improvements:

Add missing frontmatter fields
Enhance description/summary
Add usage examples
Fix formatting issues
Add cross-references

Stage 6: Index to Database

# Run component indexer to add to searchable SQLite database
python3 ~/.coditect/scripts/component-indexer.py -i  # Incremental update

# Verify component is indexed
python3 ~/.coditect/scripts/component-indexer.py --search "<component-name>"

# Show updated stats
python3 ~/.coditect/scripts/component-indexer.py --stats

Output: Component added to ~/PROJECTS/.coditect-data/context-storage/platform.db (ADR-118 Tier 1) with FTS5 search

Indexed Fields:

id - Type-prefixed unique ID (agent/my-agent, skill/my-skill)
type - Component type (agent, skill, command, script, hook)
name - Component name
path - File path
description - Auto-extracted description
model - LLM binding from frontmatter (if present)
tags - Classification tags
status - Activation status

Verification:

# Search for newly indexed component
python3 ~/.coditect/scripts/component-indexer.py --search "my-new-agent"

# Filter by type
python3 ~/.coditect/scripts/component-indexer.py --type agent --search "my-new-agent"

# Check database directly (ADR-118: platform.db for component data)
sqlite3 ~/PROJECTS/.coditect-data/context-storage/platform.db "SELECT id, name, type FROM component_index WHERE name LIKE '%my-new-agent%'"

Stage 7: Git Commit

# Stage changes
git add path/to/component.md

# Commit with descriptive message
git commit -m "feat(component): Activate <component-name>

- Registered in component index
- MoE classified (confidence: 0.95)
- Quality improved
- Indexed in SQLite database

Co-Authored-By: Claude <noreply@anthropic.com>"

# Push (optional, with --push flag)
git push origin HEAD

Full Workflow Example

# Activate a new agent
/component-activate agents/my-new-agent.md

# Output:
# Stage 1: Inventory
#   Found 1 component at agents/my-new-agent.md
#   Type: agent
#   Status: draft → active
#
# Stage 2: Register
#   Added to framework registry: agent/my-new-agent
#   Total components: 153 agents
#
# Stage 3: Activate
#   Frontmatter: ✅ Complete
#   Directory: ✅ agents/
#   Registered: ✅ Yes
#
# Stage 4: Classify
#   Running MoE classification...
#   Confidence: 0.95
#   Tags: [agent, automation, development]
#
# Stage 5: Improve
#   Added: usage examples, cross-references
#   Quality score: 85 → 92
#
# Stage 6: Index to Database
#   Running component-indexer.py -i...
#   Indexed: agent/my-new-agent
#   Database: ~/PROJECTS/.coditect-data/context-storage/platform.db (ADR-118 Tier 1)
#   Searchable: ✅ Yes (FTS5)
#
# Stage 7: Commit
#   Committed: feat(agent): Activate my-new-agent
#   Ready to push with: git push

Batch Activation

# Activate all new agents
/component-activate agents/ --new-only

# Activate all components needing classification
/component-activate . --unclassified

# Activate and push
/component-activate skills/new-skill/ --commit "Add new skill" --push

Options

Option	Description
`--inventory`	Only run inventory stage
`--register`	Only run register stage
`--classify`	Only run classify stage
`--improve`	Only run improve stage
`--index`	Only run index to database stage
`--commit "msg"`	Commit with message
`--push`	Also push after commit
`--no-classify`	Skip MoE classification
`--no-index`	Skip database indexing
`--no-commit`	Don't auto-commit
`--new-only`	Only process new/unindexed
`--unclassified`	Only process unclassified
`--dry-run`	Preview without changes
`-v, --verbose`	Detailed output

Component Types

Type	Directory	Example
`agent`	`agents/`	`agents/my-agent.md`
`skill`	`skills/*/`	`skills/my-skill/SKILL.md`
`command`	`commands/`	`commands/my-command.md`
`script`	`scripts/`	`scripts/my-script.py`
`hook`	`hooks/`	`hooks/my-hook.py`
`workflow`	`workflows/`	`workflows/category/my-workflow.json`
`standard`	`coditect-core-standards/`	`coditect-core-standards/MY-STANDARD.md`

Required Frontmatter

---
title: Component Name
component_type: agent|skill|command|script|hook|workflow|standard
version: 1.0.0
status: active
summary: Brief description
keywords:
- keyword1
- keyword2
tags:
- tag1
created: '2026-01-09'
updated: '2026-01-09'
moe_confidence: 0.950
moe_classified: 2026-01-09
---

/classify - MoE document classification
/component-create - Create new components
/which - Find components for tasks
/agent - Invoke CODITECT agents

Troubleshooting

Issue	Solution
"Component not found"	Check path is relative to ~/.coditect/
"Missing frontmatter"	Add required YAML frontmatter block
"Classification failed"	Check MoE classifier is configured
"Indexing failed"	Run `python3 ~/.coditect/scripts/component-indexer.py --stats` to check DB
"Not searchable"	Run `python3 ~/.coditect/scripts/component-indexer.py -i` to rebuild index
"Commit failed"	Ensure git repo is clean, check permissions

Script: ~/.coditect/scripts/component-activate.sh Indexer: ~/.coditect/scripts/component-indexer.py Database (ADR-118): ~/PROJECTS/.coditect-data/context-storage/platform.db (Tier 1 - component data) Last Updated: January 26, 2026

Usage​

System Prompt​

Stage 1: Inventory​

Stage 2: Register​

Stage 3: Activate​

Stage 4: Classify (MoE)​

Stage 5: Improve (Optional)​

Stage 6: Index to Database​

Stage 7: Git Commit​

Full Workflow Example​

Batch Activation​

Options​

Component Types​

Required Frontmatter​

Related Commands​

Troubleshooting​