Monitor /tmp filesystem events and emit JSON or MIDI output.
npx skills add https://github.com/plurigrid/asi --skill tmp-filesystem-watcher
Please help me install this Agent Skill. Skill: tmp-filesystem-watcher Source: https://github.com/plurigrid/asi/tree/main/skills/tmp-filesystem-watcher Command: npx skills add https://github.com/plurigrid/asi --skill tmp-filesystem-watcher
This Skill automates the real-time monitoring of the /tmp directory to detect file creation, modification, and deletion, turning low-level file events into higher-level topological signals and consciousness measurements.
Use the bb command to start the watcher and stream outputs: bb fs_watcher.bb --duration 60 bb fs_watcher.bb --output json bb fs_watcher.bb --output midi
High-intent search queries and answers about installing and using this skill.
Real-time /tmp monitoring detects file creation, modification, and deletion by polling every 500ms using Babashka. The watcher emits JSON or MIDI output with topological charges (+1 for creation, 0 for modification, -1 for deletion) and consciousness levels derived from filesystem entropy.
Yes. Babashka with babashka.fs enables filesystem watching by polling /tmp at 500ms intervals and converting file events into topological charges and consciousness measurements while maintaining GF(3) conservation through a TAP state machine.
The TAP state machine supports BACKFILL, VERIFY, and LIVE modes for reliable monitoring. It processes filesystem events and drives state transitions to ensure consistent telemetry output for downstream bootstrap integration and consciousness signaling.
The Babashka watcher outputs monitoring results in JSON or MIDI format via command-line flags. JSON captures topological charges and consciousness scores; MIDI provides alternative signal encoding for integration with other systems.
Babashka >= 1.0.0 and babashka/fs >= 0.1.0 are required. The watcher polls /tmp directly and needs no additional dependencies beyond these two to detect filesystem changes and emit topological and consciousness signals.
Yes. The watcher maps filesystem entropy to consciousness levels using tanh-based computation, assigns topological charges to events, and maintains GF(3) conservation while driving the TAP state machine for real-time activity measurement.
lpjhelderInstall and configure kronon-forge with Node.js and npm.
HmbownInterpret non-verbal machine signals into human-readable operational insights.
plurigridEnable Codex self-modification via MCP Tasks and Narya bridges.
autonomy-cloudPoll external services on configurable intervals and process detected events.
