constripacity avatar

Claude Replay

constripacity/claude-replay

Records every Claude Code session to local SQLite, then surfaces them through nine MCP tools plus a web dashboard and terminal UI. The hooks layer writes tool calls and checkpoints passively. The server exposes `replay_resume` to generate structured briefs from dead sessions, `replay_search` for full-text queries across your history, `replay_insights` for per-session metrics like death cause and error rate, and `replay_diff` to compare two runs side by side. Also ships `replay_export` to render sessions as standalone HTML traces. Runs on port 8766 alongside Claude Bridge. Think of it as observability over your entire session history, where native `--resume` and `/rewind` handle only the active session. Install once via `claude-replay install` and it records everything going forward.

GitHub stars
1
tools
10
transport
STDIO
env var
1
last updated
Aug 17, 2026

Tools

10 tools

Public tool metadata for what this MCP can expose to an agent.

replay_statusCurrent session summary objective, status, how it ended, event/checkpoint counts, last activity

Current session summary objective, status, how it ended, event/checkpoint counts, last activity

No parameter schema in public metadata yet.

replay_checkpointForce a checkpoint of the current session now, with an optional note

Force a checkpoint of the current session now, with an optional note

No parameter schema in public metadata yet.

replay_resumeGenerate a structured resume brief for a session (default: most recent)

Generate a structured resume brief for a session (default: most recent)

No parameter schema in public metadata yet.

replay_sessionsList recent sessions with status, model, duration, checkpoint count

List recent sessions with status, model, duration, checkpoint count

No parameter schema in public metadata yet.

replay_insightsPer-session metrics: how it ended, duration, tool calls, error rate, files touched, top tools

Per-session metrics: how it ended, duration, tool calls, error rate, files touched, top tools

No parameter schema in public metadata yet.

replay_statsCross-session analytics: tool calls, error rate, why sessions end, tool mix, per-project rollups

Cross-session analytics: tool calls, error rate, why sessions end, tool mix, per-project rollups

No parameter schema in public metadata yet.

replay_searchFull-text search across sessions with filters (tool, cause, date, project), ranked by match count

Full-text search across sessions with filters (tool, cause, date, project), ranked by match count

No parameter schema in public metadata yet.

replay_diffCompare two sessions: metric deltas + which files each touched

Compare two sessions: metric deltas + which files each touched

No parameter schema in public metadata yet.

replay_tagName a session and add/remove tags for later retrieval

Name a session and add/remove tags for later retrieval

No parameter schema in public metadata yet.

replay_exportRender a session as a self-contained trace (html / json / md) and return the path

Render a session as a self-contained trace (html / json / md) and return the path

No parameter schema in public metadata yet.

Sponsors

Configuration

CLAUDE_REPLAY_DB

SQLite database path (default: ~/.claude-replay/sessions.db). Point HTTP and stdio instances at the same path to share recorded sessions.

Readme

Claude Replay

The observability layer for Claude Code sessions — search, analytics, and visualization across every project.

PyPI CI Python License MCP


Claude Code's --resume/--continue and /rewind recover the session you're in. Claude Replay does the part they don't: it hooks passively into every session, records what happened to a local SQLite store, and makes every past session — across every project — searchable, comparable, measurable, and exportable. Full-text search, per-session insight metrics, death-cause classification (why a session ended), session diffing, a web dashboard, and a terminal UI — none of which Claude Code has natively.


View the full README on GitHub

Details

Repository
github.com/constripacity/claude-replay
Category
Search & Web Crawling
Package
claude-replay
Transport
STDIO
Registry
MCP registry, active

More from constripacity

Related Search & Web Crawling MCP servers

View all
1cyanheads avatarCollege Scorecard Mcp Servercyanheads/college-scorecard-mcp-serverSearch, compare, and analyze U.S. college data — costs, earnings, programs, and outcomes.1stars
2cyanheads avatarCyanheads Mcp Servercyanheads/cyanheads-mcp-serverFleet discovery for the cyanheads MCP ecosystem — semantic search + install snippets.1stars
3cyanheads avatarGbif Biodiversity Mcp Servercyanheads/gbif-biodiversity-mcp-serverSearch GBIF species taxonomy, occurrence records, datasets, and publishers.1stars
4cyanheads avatarNist Nvd Mcp Servercyanheads/nist-nvd-mcp-serverSearch and audit NIST NVD CVEs by keyword, severity, CWE, CISA KEV status, and CPE.1stars
5cyanheads avatarUsda Mcp Servercyanheads/usda-mcp-serverSearch foods, compare nutrients, and look up the full USDA FoodData Central database.1stars
6danielenricocahall avatarDisney Lorcana MCPdanielenricocahall/lorcana-mcpMCP server for Disney Lorcana TCG: card search, deck validation, stats, and deck import/export1stars