bastion/proactive-engine
Skill8 starsChanged 4 months agoMITView on GitHub
Proactive intelligence engine — detects inactivity, stale memories, CVEs, and temporal patterns, then generates context-aware suggestions via LLM. Replaces the deprecated bastion/proactive skill.
- Reads credentials
In this repository17 other files it gives its agents.
- bastion/agentskills-installSkill
- bastion/agentskills-publishSkill
- bastion/calendarSkill
- bastion-pet-packSkill
- bastion/channel-scaffoldSkill
- bastion/crisis-modeSkill
- bastion/life-logSkill
- mcporterSkill
- bastion/mesh-connectSkill
- bastion/mobile-connectSkill
- bastion/onboardingSkill
- bastion/output-validatorSkill
- bastion/persona-engineSkill
- bastion/self-improvingSkill
- bastion/skill-writerSkill
- bastion/weekly-reviewSkill
- bastion/weight-systemSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 37
Degraded mode: if memupalace is unavailable, engine continues without persistence
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Skill: bastion/proactive-engine
Objective
Monitor the user's interaction patterns across personas and generate proactive suggestions without requiring the user to ask.
Architecture
Layer 0 (Detectors) Layer 1 (Generators)
───────────────────── ──────────────────────
InactivityDetector ──┐
MemoryStalenessDetector──┤→ EventBus → SuggestionGenerator → memupalace
TemporalPatternDetector──┤ → LLM call
CVEDetector ──┘ WeeklySynthesizer
IntentTracker- EventBus: deduplicates events (6h default, 24h for CVE/staleness)
- SuggestionGenerator: single LLM call per cycle via OpenRouter
- Degraded mode: if memupalace is unavailable, engine continues without persistence
CLI Commands
Invoke via
exec python3 skills/proactive-engine/main.py <command>
run-cycle (every 2h)
python3 skills/proactive-engine/main.py run-cycle \
--personas '["carreira","estudos","projetos-pessoais"]' \
--skills '["bastion/life-log","bastion/guardrails"]'run-cve-check (every 24h)
python3 skills/proactive-engine/main.py run-cve-check \
--skills '["bastion/life-log","bastion/guardrails"]'run-weekly
python3 skills/proactive-engine/main.py run-weekly \
--personas '["carreira","estudos"]'Environment Variables
| Variable | Default | Description |
PROACTIVE_ENABLED | true | Set to false to disable entirely |
PROACTIVE_LLM_MODEL | google/gemini-flash-1.5 | OpenRouter model |
PROACTIVE_INACTIVITY_DAYS | 3 | Days before marking persona inactive |
PROACTIVE_STALENESS_DAYS | 14 | Days before marking memory stale |
PROACTIVE_LIFELOG_WINDOW | 50 | Records to include in LLM context |
PROACTIVE_DEDUP_WINDOW_HOURS | 6 | Event dedup window (hours) |
BASTION_SKILL_REGISTRY_URL | (empty) | Optional compatible skill-vulnerability registry; CVE checks are disabled if unset |
BASTION_SKILL_REGISTRY_TOKEN | (empty) | Bearer token for that registry |
OPENROUTER_API_KEY | (required) | OpenRouter API key for LLM calls |
Persistence
| File | Purpose |
db/proactive-engine/pending-events.json | Unprocessed detection events |
db/proactive-engine/intent-queue.json | Intent queue (when memupalace offline) |
db/proactive-engine/heartbeat-state.json | Last run timestamps per task |
Memupalace Wings
| Wing | Content |
proactive/suggestions | Generated suggestions |
proactive/intent | Tracked user intents |
proactive/weekly | Weekly summaries |
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.