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 hamid-bhz/claude-skills --skill kill-stale-mcpgit clone --depth 1 https://github.com/hamid-bhz/claude-skillsWrote 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/hamid-bhz/claude-skills/kill-stale-mcp)<a href="https://agentmods.dev/skills/hamid-bhz/claude-skills/kill-stale-mcp"><img src="https://agentmods.dev/badge/skills/hamid-bhz/claude-skills/kill-stale-mcp/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/hamid-bhz/claude-skills/kill-stale-mcp"><img src="https://agentmods.dev/badge/skills/hamid-bhz/claude-skills/kill-stale-mcp.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00071 | $0.00650 |
| Opus 5 | $0.00036 | $0.00325 |
| Sonnet 5 | $0.00014 | $0.00130 |
| Haiku 4.5 | $0.00007 | $0.00065 |
Grade A, and why
kill-stale-mcp 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 11d 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.
How it starts
The opening of the file, as written. The whole thing — 41 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/kill-stale-mcp
npx-launched MCP servers leak. The launch chain is npm exec <server>@... → server → (sometimes) a telemetry watchdog. When a session ends uncleanly, the termination signal stops at the npm exec wrapper — npx does not forward it to its children — so the whole group keeps running. Every dead session leaves one group behind (~100–250 MB each).
This skill kills the stale groups and keeps the live ones.
Staleness test: a live group's npm exec wrapper is parented to a running host process (Claude Code, an IDE extension host). When the host dies, the orphaned processes are re-parented to PID 1 (launchd on macOS, init/systemd on Linux). PPID 1 → stale. PPID = a live host → in use, keep.
Steps
-
List every MCP-related process with its parent PID:
ps -eo pid,ppid,lstart,command | grep -i mcp | grep -v grep -
Classify each process:
- Not ours — never touch: anything owned by an IDE or app rather than launched via npx (e.g. Cursor
mcp-process, GitLens/GitKrakengk mcp, Electron app helpers). Only handle processes whose command containsnpm execor a~/.npm/_npx/path, plus their direct children. - Stale: an npx-launched process with
PPID 1; or a watchdog whose--parent-pid=<N>is dead (check withkill -0 <N>). - Live: the wrapper's PPID is a running session process. Keep the whole group.
- Not ours — never touch: anything owned by an IDE or app rather than launched via npx (e.g. Cursor
-
Kill the stale PIDs explicitly (never
pkill -f mcp— that would take down live sessions' servers and IDE-owned processes):kill <stale pids...> -
Re-run the listing from step 1. Killing a wrapper re-parents its children to PID 1, so they now show up as stale themselves — repeat step 3 until no npx-launched process with PPID 1 remains.
-
Report: how many groups were killed (with spawn times) and which live groups were kept.
Notes
- Live servers respawn on demand the next time a session uses them — this skill removes leaked instances, it does not disable any MCP.
- Root cause is npx signal handling. If one server leaks constantly, launch it directly by its bin path in the MCP config instead of via
npx.
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.
- 11d ago First seen · 41 lines · 71 tokens per session scan A 9bef98cd3a9b
kill-stale-mcp is a skill published in the GitHub repository hamid-bhz/claude-skills (3 stars, last pushed 1mo ago), licensed MIT. It adds 71 tokens to every session and 650 once invoked, about $0.0004 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
claude-md-improver
Audit and improve CLAUDE.md files in repositories. Use when user asks to check, audit, update, improve, or fix CLAUDE.md files. Scans for all CLAUDE.md files, evaluates quality against templates, outputs quality report, then makes targeted updates. Also use when the user mentions "CLAUDE.md maintenance" or "project…
agent-platform-rag-engine-management
Manage and query Agent Platform RAG Engine Corpora and retrieve grounded contexts using the Google GenAI SDK. Use when listing RAG corpora or files, inspecting a corpus, retrieving contexts, or generating content grounded in a RAG corpus. Do not use for standard database queries (use SQL/Spanner skills), Google…
gke-reliability
Improves GKE workload reliability, using PDBs, health probes, and topology spread constraints. Use when configuring GKE workload reliability, setting up PDBs, or configuring GKE health probes (liveness, readiness, startup). Don't use for disaster recovery setup or full cluster backups (use gke-backup-dr instead).
gke-workload-security
Audits, configures, and hardens workload-level security controls for Google Kubernetes Engine (GKE) applications and namespaces. Covers running cluster security audits (auditcluster.sh), configuring Workload Identity Federation (impersonation, KSA/GSA binding, and pod setup), enforcing Network Policies (default-deny…
agent-platform-model-registry
Agent Platform Model Registry Management. Use when you need to upload, list, describe, update, or delete machine learning models (and their versions) in the Agent Platform Model Registry. Don't use for model training, model deployment to endpoints, or managing non-Agent Platform models.
google-cloud-solution-agentic-analytics-spark-knowledge-catalog
Discovers requirements and generates guidance to design and deploy a governed, secure agentic-analytics solution for data that's distributed across Google Cloud, other cloud providers, or on-premises. Data that's outside Google Cloud (such as data from Databricks, Snowflake, Salesforce, SAP, or Oracle systems) is…