Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add NeoZhouCHN/github-stars-radar --skill github-stars-radargit clone --depth 1 https://github.com/NeoZhouCHN/github-stars-radarWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/neozhouchn/github-stars-radar/github-stars-radar)<a href="https://agentmods.dev/skills/neozhouchn/github-stars-radar/github-stars-radar"><img src="https://agentmods.dev/badge/skills/neozhouchn/github-stars-radar/github-stars-radar.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00029 | $0.00448 |
| Opus 5 | $0.00015 | $0.00224 |
| Sonnet 5 | $0.00006 | $0.00090 |
| Haiku 4.5 | $0.00003 | $0.00045 |
Grade A, and why
github-stars-radar scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
GitHub Stars Radar
Use this skill when a task may benefit from the user's own starred repositories: library selection, MCP/plugin discovery, prompt/tool examples, reference implementations, or prior art for Codex, Claude Code, OpenClaw, and Hermes.
When To Search Local Stars First
- The user asks for reusable projects, libraries, plugins, MCP servers, or example repos.
- The task resembles something the user may have already saved for later.
- You need a shortlist of candidates before broader web research.
- You want to avoid repeating analysis already saved in local agent memory.
Call search_stars for keyword lookup, recommend_stars_for_task for task-based ranking, and get_star for one known repo. Use sync_stars for a fast metadata refresh; use sync_readmes only when README text is needed, and run it in small batches.
When External Research Is Still Required
For major route decisions, do not rely only on local stars. You must search the web and compare at least 3 mature external projects when the decision affects:
- architecture shape
- repository structure
- commercialization or distribution
- license
- payment or account systems
- deployment, compliance, or security route
Local stars are a memory aid, not proof of current best practice.
Output Shape
When recommending repos, include:
- repository name
- score
- top score signals from
score_breakdown - why it fits
- why it may not fit
- cost or risk
- next verification step
Use the numeric score from recommend_stars_for_task directly. Summarize only the strongest 2-3 nonzero score_breakdown signals, such as readme_match, topic_match, popularity, recency, or analysis_quality; do not paste the whole JSON unless the user asks for raw output.
If local sync fails but stale cache is available, keep working and disclose the stale-cache status from the tool result.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 8d ago First seen · 47 lines · 29 tokens per session scan A 6da872692f7c
github-stars-radar is a skill published in the GitHub repository NeoZhouCHN/github-stars-radar (5 stars, last pushed 3mo ago), licensed MIT. It adds 29 tokens to every session and 448 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
using-goodmemory
Recall and persist durable project context with GoodMemory in Kimi Code.
turritopsis-agent-memory
Design, review, or repair Turritopsis Stages for Agent memory systems, shared project memory, long-term recall, evidence ingestion, retrieval indexes, corrections, retention, and memory maintenance. Use when a project stores knowledge across sessions or agents and must keep raw evidence, canonical truth, summaries…
turritopsis-personal-ai-assistant
Design, review, or repair Turritopsis Stages for a private or personal AI assistant, including human-assistant roles, consent and authority, proactive behavior, tool actions, relationship continuity, private channels, memory correction, visible explanations, and human direction. Use when the system's value depends on…
chain
Inject the 3 most recent session IDs for the current workspace into the prompt before Claude responds. Use when starting work and wanting recent session context.
repo-brain
Use Repo Brain to locate current repository code, consult evidence-backed memory, log reads, detect stale analyses, and refresh memory after changes.
repo-brain
Use Repo Brain to locate current repository code, check stale memory, log reads, and refresh memory after edits.