Display Name
sivtr
Category
Slash-Commands
Sub-Category
Slash-Commands: Project & Task Management
Primary Link
https://github.com/Ariestar/sivtr
Author Name
Ariestar
Author Link
https://github.com/Ariestar
License
Apache-2.0
Other License
No response
Description
Rust CLI + agent skill that unifies terminal output and AI session history into a searchable, reusable memory layer
for humans and agents.
Validate Claims
Install with cargo install sivtr and npx skills add Ariestar/sivtr --skill sivtr-memory -g. Then run sivtr in any project — you get a vim-style TUI to browse, search, and select output. Ask Claude Code "what
was the last test failure?" with the skill installed — it will query sivtr's local history instead of re-running tests.
Specific Task(s)
Run a few terminal commands that produce long output (e.g., cargo test, npm run build), then use the skill to ask Claude Code to
find specific information from previous command output without re-running anything or pasting any information. For example, after running tests
that fail, ask Claude Code to find the failing test name and error message from the terminal history — not by re-running
the tests, but by reading sivtr's captured output.
Specific Prompt(s)
use the sivtr slill and ask claude to find waht happened in the terminal.
Additional Comments
No response
Recommendation Checklist
Display Name
sivtr
Category
Slash-Commands
Sub-Category
Slash-Commands: Project & Task Management
Primary Link
https://github.com/Ariestar/sivtr
Author Name
Ariestar
Author Link
https://github.com/Ariestar
License
Apache-2.0
Other License
No response
Description
Rust CLI + agent skill that unifies terminal output and AI session history into a searchable, reusable memory layer
for humans and agents.
Validate Claims
Install with
cargo install sivtrandnpx skills add Ariestar/sivtr --skill sivtr-memory -g. Then runsivtrin any project — you get a vim-style TUI to browse, search, and select output. Ask Claude Code "whatwas the last test failure?" with the skill installed — it will query sivtr's local history instead of re-running tests.
Specific Task(s)
Run a few terminal commands that produce long output (e.g.,
cargo test,npm run build), then use the skill to ask Claude Code tofind specific information from previous command output without re-running anything or pasting any information. For example, after running tests
that fail, ask Claude Code to find the failing test name and error message from the terminal history — not by re-running
the tests, but by reading sivtr's captured output.
Specific Prompt(s)
use the sivtr slill and ask claude to find waht happened in the terminal.
Additional Comments
No response
Recommendation Checklist