slacrawl

Why it's ranked here

The breadth is unusually practical: several ingestion routes converge on one archive, while search, diagnostics, reporting, automation output, retention, and sharing cover the archive lifecycle. The main reservations are token-dependent coverage, platform limits, and the absence of working embedding support.

What's good

SQLite FTS5 search includes a substring fallback. Text, JSON, and log output serve both people and scripts. Purging previews changes by default, keeps threads together, and can remove cached media. Shared snapshots merge without deleting destination-only rows unless exact restoration is explicitly requested.

Tradeoffs

Broader thread and direct-message coverage can require a user token, while live tailing requires an app token. Desktop-cache recovery supports macOS and Linux only. Homebrew may lag releases. One configuration and database form a visibility boundary, so personal and company archives need deliberate separation.

How to use it well

It suits engineers, operators, and researchers who want recurring local search, incident lookup, reports, or controlled archive sharing. Start with an export or scoped API sync, verify coverage through diagnostics, then automate incremental refreshes and structured output. It does not provide semantic embedding search.

Observed

License

MIT

Primary language

Go

Interface

Command-line application with a terminal browser, structured commands, and read-only SQL

Install surface

Homebrew, release archives, Debian and RPM packages, source builds, and Docker container builds

Platform support

macOS and Linux; release artifacts target AMD64 and ARM64

Ingestion interfaces

Slack API, Slack Desktop caches, MCP connector, trusted JSONL provider, and Slack exports

Automation output

Text, JSON, and log formats, plus bash and zsh completion