Score8.3
Popularity82.0
Risknone
TierGold
Score breakdown
Usefulness9.0
Novelty8.0
Momentum7.0
Maturity8.2
Open-source/build8.4
Evidence8.0
Workflow potential9.4
Setup ease8.8
Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.
Why it matters
Useful for Claude Code power users who hit the **context-window visibility problem** — Claude Code ships a 200k-token window by default and there is no built-in way to see the running total short of opening the JSONL log files, so a user who keeps losing sessions to context overflow has no signal until the model starts dropping turns; the dashboard shows a percentage bar with per-turn breakdown so
Who should use it
Claude Code power users who hit the **context-window visibility problem** — Claude Code ships a 200k-token window by default and there is no built-in way to see the running total short of opening the JSONL log files, so a user who keeps losing sessions to context overflow has no signal until the model starts dropping turns; the dashboard shows a percentage bar with per-turn breakdown so the user can see which prompts blew up the context and course-correctUsers who have **lost sessions to Ctrl-C / terminal crash / laptop sleep** — every Claude Code prompt and response is logged locally to a JSONL append-only file at `~/.claude-pulse/sessions/<session_id>.jsonl`, the dashboard parses that log to render a per-session timeline, and the user can scroll back through everything Claude Code said even if the terminal diedUsers who need **full-text search across all session transcripts** for audit (e.g. searching for `Bearer `, `sk-`, `password=` to catch accidental secret leaks in a transcript) — the dashboard indexes the entire JSONL log directory at startup with subprocess `grep` for speed and no index file, so the search is always liveUsers who want **approve-tool-call-from-your-phone** mode — the dashboard exposes a small HTTPS server with self-signed cert on the LAN, a Claude Code session that hits a destructive tool (`rm -rf`, `git push --force`, AWS `terminate-instances`, etc.) can pause for an approval gate, the user gets a notification on the same-network phone, opens the dashboard, taps Approve or Reject, and the agent resumesUsers who hate **dependency hell** — the dashboard is `import sys, os, json, http.server, ssl, subprocess, pathlib, datetime` and a single inlined HTML file with vanilla JS (no React, no Vue, no build step, no `requirements.txt`, no virtualenv), runs on a fresh Python 3.10+ install with no setup beyond `python claude-pulse.py`Users who care about **local trust boundary** — the HTTPS server binds to the user's LAN IP, the self-signed cert is generated at first run, there is no cloud relay, no telemetry, no auth service (the approval gate uses LAN-network proximity as the auth, appropriate for a household / office network)Evaluation: `git clone https://github.com/nikitadoudikov/claude-pulse && cd claude-pulse && python claude-pulse.py` and open `localhost:7645`; for the mobile setup, add `--host 0.0.0.0` to expose on the LAN
Who should skip it
Pass on nikitadoudikov/claude-pulse if its scope or audience does not match what your team is building right now.
About this signal
nikitadoudikov/claude-pulse is tracked by RepoRadar as a zero-dependency local dashboard in the MIT zero-dependency (just Python 3 stdlib + 1 ht section. It was first seen on 2026-06-25 and last updated on 2026-06-25. The current verdict is 'try now' with a Gold tier and easy setup difficulty. nikitadoudikov/claude-pulse leads on workflow potential (9.4) and practical usefulness (9.0); its lowest signal is momentum (7.0), so factor that in before investing setup time. This page summarizes the evidence RepoRadar has captured from captured source metadata. The score, tier, risk label, and verdict on this page are never influenced by sponsorship, ads, or tips — they reflect only the usefulness, popularity, novelty, momentum, maturity, and evidence signals described in the RepoRadar methodology.
How this item is evaluated
RepoRadar assigned nikitadoudikov/claude-pulse a composite score of 8.3 out of 10, placing it in the Gold tier. This score combines weighted sub-signals: usefulness (35%), novelty (18%), momentum (14%), maturity (10%), open-source/build quality (7%), evidence quality (6%), workflow potential (6%), and setup ease (4%). Popularity is tracked separately at 82.0 and never affects the composite score or tier. The risk label of 'none' reflects inherent user-impacting hazards, not generic novelty. Items with no risk flag may still require normal code review before production use.
Risk explanation
No inherent user-impacting risk is flagged from the captured evidence.
Evidence links
Closest alternatives / related signals
claude-pulsenikitadoudikovclaude-codeclaude-code-dashboardzero-dependencypython-stdlibsingle-python-filesingle-html-file
