#394394th of 455 in Skills, plugins and prompts
ctx
Local CLI to search agent session history and blame code to sessions
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 1.2k
- License
- Apache-2.0
- Last commit
- Nov 2026
- Last release
- Nov 2026
- Language
- Rust
Missed the latest GitHub check; these facts may be older.
Overview
ctx indexes coding agent transcripts already on disk (Claude Code, Codex, Cursor, Pi, Gemini CLI and others) into a local Tantivy index, with BM25 search, optional local semantic search and cited retrieval. It also provides `ctx blame` to trace commits to the agent session that wrote them, `ctx graph` for a SQLite-backed code graph, and `ctx sift` to compact noisy tool output.
Who it is for: Developers running coding agents who want searchable, local session history
Strengths
- Reads existing agent logs without hooks or code inside the agent process
- Single Rust binary; search, blame and graph run locally with no API key
- ctx graph opens existing Graphify .graf/index.db files without conversion
- Installs as a skill or MCP server for Codex, Claude Code and Cursor
Weaknesses
- Transcripts are indexed unredacted, so copied output needs manual review
- Blame cannot attribute code from sessions not on your machine
- Team history backup and sharing is beta and needs a ctx server you run
- Speed and token benchmarks are the authors' own, on their workloads
What it needs
- no GPU
- Models: Claude Code, Codex, Cursor, Pi, Gemini CLI
Also in Skills, plugins and prompts
See all 455| Rank | Project | Score |
|---|---|---|
| 1 | claude-memPersistent cross-session memory plugin for Claude Code and other coding agents | 97 out of 100 |
| 2 | impeccableDesign skill with 24 slash commands for AI coding agents' frontend output | 97 out of 100 |
| 3 | gstackMarkdown slash-command skills that give Claude Code planning, review, QA and ship roles | 96 out of 100 |
| #4 | rufloAgent harness adding swarms, memory and hooks to Claude Code and Codex | 96 out of 100 |
| #5 | cavemanShortens coding-agent replies and compresses tool output before the model reads it | 95 out of 100 |