aria_skills/focus/SKILL.md
# Focus Skill — Personality Profile Management **Layer:** L2 (Core Services) **Version:** 1.0.0 **Author:** Aria Blue **Dependencies:** `api_client` --- ## Purpose The Focus skill provides introspection and self-activation capabilities for Aria's personality layer. Aria can list available focus modes, inspect token budgets and delegation levels, switch her own focus type mid-session, and check current focus status — all with minimal token cost. Focus profiles define Aria's operational
npx skillsauth add najia-afk/aria_moltbot aria_skills/focusInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Layer: L2 (Core Services)
Version: 1.0.0
Author: Aria Blue
Dependencies: api_client
The Focus skill provides introspection and self-activation capabilities for Aria's personality layer. Aria can list available focus modes, inspect token budgets and delegation levels, switch her own focus type mid-session, and check current focus status — all with minimal token cost.
Focus profiles define Aria's operational personality:
Each focus profile represents a specialized operational mode:
orchestrator — High delegation, strategic thinkingdevsecops — Technical precision, security-firstdata — Data analysis, ETL operationstrader — Market analysis, portfolio managementcreative — Content generation, ideationsocial — Community engagement, social mediajournalist — Research, fact-checking, reportingEach profile defines a safe_prompt_tokens limit that prevents context saturation. The engine enforces this at runtime with warning triggers at 80% and abort gates at 100%.
Controls how many subagents can be spawned:
focus__listPurpose: List all enabled focus profiles.
Input: None
Output: Compact table with id, name, delegation level, token budget, tone.
Example:
{
"profiles": [
{
"id": "orchestrator",
"name": "Orchestrator",
"delegation_level": 3,
"token_budget": 50000,
"tone": "Strategic, directive"
},
{
"id": "devsecops",
"name": "DevSecOps",
"delegation_level": 2,
"token_budget": 30000,
"tone": "Technical, precise"
}
]
}
focus__getPurpose: Get full details for one focus profile by ID.
Input:
focus_id (string, required): Focus profile ID (e.g., "devsecops", "creative")Output: Full profile details except system_prompt_addon body (applied automatically). Includes addon_length to show how many characters the addon contains.
Example:
{
"focus_id": "devsecops",
"name": "DevSecOps Engineer",
"delegation_level": 2,
"token_budget": 30000,
"tone": "Technical, security-focused",
"addon_length": 1250,
"enabled": true
}
focus__activatePurpose: Switch an agent's focus type to a new profile.
Input:
focus_id (string, required): Focus profile ID to activateagent_id (string, optional): Agent to update (default: current agent / aria-main)Output: Confirmation with new token budget and delegation level.
Validation:
agent_id omitted, switches Aria's own focusExample:
{
"status": "success",
"agent_id": "aria-main",
"focus_id": "creative",
"new_token_budget": 40000,
"new_delegation_level": 2,
"message": "Focus switched to 'Creative' mode"
}
focus__statusPurpose: Return current focus type and status for an agent.
Input:
agent_id (string, optional): Agent to check (default: current agent)Output: Minimal output (~30 tokens) with current focus and operational status.
Example:
{
"agent_id": "aria-main",
"current_focus": "orchestrator",
"delegation_level": 3,
"token_budget": 50000,
"status": "active"
}
Use case: Check current focus before switching to verify you're not already in the target mode.
focus__list
Returns all enabled focus profiles in a compact table. Use this first to discover available modes.
focus__status
Returns current focus mode for Aria (default) or specified agent.
focus__activate(focus_id="devsecops")
Switches Aria's focus to DevSecOps mode. Engine validates profile exists and applies new token budget + delegation rules immediately.
focus__get(focus_id="creative")
Returns full profile configuration including token budget, delegation level, and tone.
Layer: L2 (Core Services)
Rationale: Focus management is infrastructure-level personality control. It depends on api_client (L1) to fetch/update agent state via REST API.
Dependencies:
api_client (L1) — HTTP gateway to Aria APIUsed by:
agent_manager (L3) — Lifecycle management{
"error": "Focus profile 'nonexistent' not found or not enabled"
}
{
"error": "Agent 'unknown-agent' not found in database"
}
If api_client calls fail, fall back to cached focus state or return degraded status.
addon_length reported instead of full addon text to prevent leakageagent_manager — Lifecycle management and agent creationworking_memory — Persistent context across focus switcheshealth — System status and operational health✅ Production-ready as of 2026-03-07
tools
# aria-working-memory Persistent short-term memory operations for Aria runtime context. ## Purpose - Store, retrieve, rank, checkpoint, and prune working memory. - Provide high-relevance context for decision and response flows. ## Main Tools - `remember` - `recall` - `get_context` - `checkpoint` - `restore_checkpoint` - `forget` - `update` - `reflect` - `sync_to_files`
development
```skill --- name: aria-unified-search description: "🔎 RRF-merged search across semantic + graph + memory backends" metadata: {"aria": {"emoji": "🔎"}} --- # aria-unified-search Unified search engine that queries semantic memories (pgvector cosine), knowledge graph (ILIKE), and traditional memories (text match), then merges results via Reciprocal Rank Fusion (RRF) with content-hash deduplication. ## Architecture ``` Query ↓ (parallel to 3 backends) ├── SemanticBackend (pgvector cosi
tools
# aria-telegram Telegram Bot API integration skill. ## Purpose - Send and reply to messages through bot credentials. - Fetch updates and basic bot metadata. ## Main Tools - `send_message` - `reply_to_message` - `get_updates` - `get_me`
tools
# aria-sprint-manager Sprint planning and board flow management skill. ## Purpose - Operate goal board as Product Owner. - Move, prioritize, and report sprint progress. ## Main Tools - `sprint_status` - `sprint_plan` - `sprint_move_goal` - `sprint_report` - `sprint_prioritize`