mirror-log is a local-first event log for capturing notes, commands, snippets, and other context you do not want to lose. Persisted events live in SQLite. Staged events live as JSON files in staging/ until you review or process them.

Version 0.1.9 reflects the current shipped behavior: a lean SQLite core, staging helpers for review-oriented workflows, an attention/decay layer, and optional embedding or inference integrations behind Cargo features.

What It Does

• Persists events to SQLite with UUIDs, timestamps, hashes, and chunk records
• Stages reviewable JSON events in staging/
• Searches full events and chunked content
• Tracks duplicates by content_hash
• Exposes attention and decay views over persisted events
• Supports optional embeddings, inference backends, iteration tables, and clipboard ingestion

Current Workflow

# Stage a single event as JSON in ./staging
mirror-log add "Overhead allocation needs review" --source journal

# Stage a file as JSON in ./staging
mirror-log add-file notes.md --source meetings

# Persist stdin events to SQLite in batches, then also write staged copies
cat ideas.txt | mirror-log stdin --source ideas

# Review staged events
mirror-log review

# Detect simple patterns from staged events
mirror-log infer

# Render staged events back out
mirror-log regenerate --output human.md

# Query persisted events
mirror-log show --last 10
mirror-log search "allocation"
mirror-log search "allocation" --chunks
mirror-log get <event-id>

# Database and integrity views
mirror-log stats
mirror-log info
mirror-log verify

# Attention layer
mirror-log attention
mirror-log attention --flagged
mirror-log attention --stats
mirror-log add-to-attention <event-id>

Installation

git clone https://github.com/CromboJambo/mirror-log
cd mirror-log
cargo build --release

# Optional feature paths
cargo build --release --features embedding
cargo build --release --features inference
cargo build --release --features clipboard

# Or install locally
cargo install --path .

The binary is written to target/release/mirror-log.

Storage Model

Persisted log data lives in mirror.db by default.

• events stores the canonical persisted event log
• chunks stores chunked slices for large content and chunk search
• decay and shadow_state support attention/visibility behavior
• event_embeddings, event_tags, event_links, and iteration tables support enrichment-oriented workflows

Staged data is separate from the persisted log.

• add and add-file write StagedEvent JSON files to staging/
• stdin currently persists events to SQLite first and then emits staged copies for review
• Persisted events are append-only; staged files are a working set

Features

The default build is intentionally lean.

• embedding: tokenizers-backed embedding generation and similarity search
• inference: HTTP inference backend support
• iteration: iteration query and type exports
• clipboard: clipboard watcher support

Attention and staging are part of the current default build, not separate feature flags.

Documentation

• USER_GUIDE.md: usage, storage, and workflow details
• src/lib.rs: public module and re-export surface
• CHANGELOG.md: release history

Direct SQLite Access

SELECT datetime(timestamp, 'unixepoch'), source, content
FROM events
ORDER BY timestamp DESC
LIMIT 10;

SELECT source, COUNT(*)
FROM events
GROUP BY source
ORDER BY COUNT(*) DESC;

SELECT COUNT(*) AS total,
       COUNT(DISTINCT content_hash) AS unique_events
FROM events;

Development

cargo fmt
cargo clippy --all-targets --all-features -- -D warnings
cargo test

Dependency audit helpers:

scripts/dep-audit.sh
scripts/dep-audit.sh --features embedding
scripts/log-dep-audit.sh
scripts/log-dep-audit-matrix.sh

License

AGPL-3.0-or-later. See LICENSE.