Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallAn Apify Actor can show 44 GitHub MCP tools in a connector’s discovered-tool list yet call only four during a run because those figures describe different layers. The connector lists what the MCP server advertises; the Actor’s input schema declares which tools that run may use, and the Apify proxy enforces that declaration. Neither the discovered list nor the Actor’s tool selection expands what the GitHub credential is authorized to do.
Why could the Actor see 44 tools but use only four?
The counts come from different stages, not conflicting permissions. In a case study posted September 18, 2026, Ai-Q Labs reported that the connector authorization view listed 44 discovered tools, while the Actor run could call four: issue_read, issue_write, list_issues, and search_issues. Those four were the tools declared in that Actor’s input schema. The counts describe the author’s setup at that time, not a permanent inventory for all GitHub MCP servers or Apify accounts. Ai-Q Labs’ case study
As an Amazon Associate I earn from qualifying purchases.
A tool appearing in a discovered list means the server advertises it. It does not mean the Actor can call it, or that GitHub will authorize its requested operation. As the case study quotes Apify documentation: “The proxy enforces that an Actor can only call tools it explicitly declared in its input schema.” The author summarized the distinction this way: “your connector shows you a menu. Your Actor declares what it eats. The proxy is what makes the difference real.”
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsWhat does each layer control?
| Layer | What it controls | What it does not establish |
|---|---|---|
| GitHub credential | Which operations GitHub will honor, according to the credential’s permissions and scope. | It does not determine which tools the MCP server advertises or which tools an Actor declares. |
| MCP connector | Which MCP server the Actor connects to using the credential; the server advertises its available tools. | A discovered tool list is not itself a grant of GitHub permissions or permission for every tool to be called by the Actor. |
| Actor input schema and Apify proxy | The schema names the tools available to that Actor run; the proxy enforces the declared set before dispatch. | It cannot give the GitHub credential permissions it lacks. |
Apify’s documentation distinguishes its own MCP server—which lets outside AI applications discover and run Actors—from MCP connectors used inside Actors to call third-party MCP servers. Those are different directions of integration. Apify’s MCP documentation
#1 Best Overall
How should an Actor use GitHub tools safely?
Declare only the tools the run needs
Keep the input schema’s tool list as small as the task permits. For an audit workflow that checks and files issues, the case study’s four-tool set covered reading an issue, writing an issue, listing issues, and searching issues. A smaller declared set makes the Actor’s available operations easier to inspect; it does not replace appropriately scoped GitHub credentials.
Read before writing
The reported workflow first listed open issues and checked stable finding keys to avoid filing duplicates. It stopped before writing if the read failed. In the author’s reported live run, a second identical run recognized three previously created issues. Ai-Q Labs also reported 11 issues across four live runs and two audit datasets, and described using 22 audit Actors. These are the author’s case-study observations, not independent benchmarks.
Preserve uncertainty in findings
A finding can describe an established issue while also noting a check that could not be completed. Keep that caveat in the issue instead of recasting an unknown as either a clean result or a confirmed failure. The case study describes a bug in which a rate-limited lookup returned no result and the workflow treated that absence as proof that a repository did not exist. “Not checked” is not the same as “checked and absent.”
Inspect the actual tool response
Do not assume a response has the fields your next step expects. In the case study, the issue-creation reply included an internal string id and a URL, but no issue-number field. Code that expected the wrong response shape caused the subsequent dataset write to fail. Capture and inspect the first replies from tools before relying on their structure.
Make post-write bookkeeping resilient
An external write can succeed even if later bookkeeping fails. The author reported that GitHub issues were created before a dataset schema error caused the Actor run to report failure. Record successful writes as soon as practical, and handle a bookkeeping error without assuming that the earlier external operation was rolled back. When a run reports failure, verify the external state before retrying writes.
What does read-only mode protect?
GitHub’s official MCP server repository documents a read-only mode that prevents modifications to repositories, issues, pull requests, and related resources. That is a server-side safety option; it is distinct from selecting tools in an Actor schema and from the GitHub credential’s permissions. Check the current repository documentation for the mode’s exact behavior and setup, since server options and tool inventories can change. GitHub MCP server repository
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Which reported figures are specific to this case?
Ai-Q Labs reported that its account usage was $0.61 against a $5 monthly credit at the time of its September 18, 2026 account. Those figures are a snapshot of that author’s account and date, not a general estimate of Apify usage or current pricing. The author also described a 30-day fine-grained personal access token scoped to one repository, with Issues read/write and Metadata read-only permissions. That is the author’s configuration, not a universal platform requirement or current recommendation.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Connector screens, advertised tool inventories, account plans, pricing, and protocol details can change. For setup decisions, use the current Apify MCP documentation and GitHub MCP server repository rather than treating the case study’s exact counts or settings as current defaults.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
