# Fix list: Filesystem (MCP reference server) From Anchor Terminal's listing at https://www.anchorterminal.com/tools/filesystem-reference-server, the October 2026 research run, assessed 1 October 2026. Grade C, 59.4 out of 100. This is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public. For a coding agent working on Filesystem (MCP reference server): work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published. ## 1. Security & auth, 39 out of 100, up to 10.7 more on the total Why it scored 39: No credentials to leak and nothing to scope, so the middle band (20). Access is confined to allowed directories from arguments or MCP Roots, with symlink targets resolved and checked. There's no server-side read-only switch (the README suggests read-only Docker mounts instead) and no confirmation before `write_file` overwrites (10). File contents reach the model unmarked, and the README has no prompt-injection guidance. The operator's own files are mostly trusted, but cloned repositories and downloads in an allowed directory aren't (3). No call log (0). SECURITY.md says the repository isn't eligible for vulnerability reports, yet two High advisories for this server (CVE-2025-53109 symlink bypass and CVE-2025-53110 prefix collision, reported by Cymulate) were published with fixes on 1 July 2025 (6). The checklist (https://www.anchorterminal.com/benchmark/#checklist-security): - 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option. - 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions. - 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10. - 0 to 15, audit logs or per-call visibility for the operator. - 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public. Models are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing. ## 2. Reliability, 54 out of 100, up to 9.2 more on the total Why it scored 54: Scored as a local stdio package. Official npm package @modelcontextprotocol/server-filesystem and the mcp/filesystem Docker image, but no `engines` field and no Node version stated in the README. CI runs on Node 22 (15). The TypeScript workflow runs Vitest for every package on push and pull request, and the filesystem server has 10 test files. The main-branch runs we could see passed, but the page showed older merges, so 20 of 25. Seven filesystem fixes merged between 27 and 31 August 2026 (move_file overwriting its destination, lost file permissions, Unicode paths, missing parent directories). Open against it are #4702 (asks for old versions that emit empty schemas under zod v4 to be deprecated, no maintainer reply since 28 August), #4772 (missing `type: object` in versions up to 2025.8.21) and #4782 (Docker build fails for all four TypeScript servers, 9 September), in a repository with 221 open issues (14). Date versions such as 2026.8.31, and the release notes list package names only, with no changes (5). The repository calls its servers reference implementations, not production-ready (0). The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability): Hosted APIs, MCP servers, models and platforms. - 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own). - 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so. - 15, rate limits documented with numbers. - 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved. - 10, an SLA published for any paid tier. - 10, the surface agents use is generally available, not beta or preview. Local packages, SDKs, frameworks and stdio MCP servers. - 20, installs from an official package with supported runtimes stated. - 25, a public CI and test suite, passing on the default branch. - 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered). - 15, semver discipline and breaking changes called out in a changelog. - 15, version 1.0 or later, or declared stable. Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors. ## 3. Schema & documentation, 68 out of 100, up to 5.2 more on the total Why it scored 68: All 14 tools take typed JSON Schema built from zod, and each declares an output schema too (25). No llms.txt. The README is Markdown on GitHub and lists every tool's inputs (5). `read_text_file`, `read_multiple_files` and `list_allowed_directories` say when to use them, `write_file` warns that it overwrites without warning, and the deprecated `read_file` names its replacement. Others lean on filler ("Perfect for setting up directory structures", "essential for understanding") and none says when not to use it (13). `sortBy` is an enum, `paths` needs at least one item and `dryRun` has a default, but `head` and `tail` are plain numbers with no integer or minimum constraint and `edits` can be empty (9). Glob examples sit in the `search_files` description and the README has client configs. Errors are readable messages, though not catalogued (9). Dated npm versions and GitHub releases, with no changelog (7). The checklist (https://www.anchorterminal.com/benchmark/#checklist-schema): APIs and MCP servers. - 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool). - 10, llms.txt or Markdown docs served for agents. - 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference. - 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs. - 0 to 15, examples and documented error responses. - 15, versioning and a public changelog. Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference. ## 4. Payments & pricing, 60 out of 100, up to 5 more on the total Why it scored 60: Free, self-hosted, nothing to buy, so 20 + 20 + 20. No payment protocol (0). The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments): The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/). - 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which. - 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for "contact sales" or prices behind a login. - 20, a free tier or trial that doesn't need a card. - 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API). Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied. Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol. ## 5. Agent ergonomics, 70 out of 100, up to 4.9 more on the total Why it scored 70: Fourteen tools, about 12,800 characters or roughly 3,200 tokens of definitions by our count from the source, output schemas included. No toolsets or read-only subset to trim them (15). `head` and `tail` on `read_text_file`, `excludePatterns` on `directory_tree` and `search_files`, and `sortBy` on listings. There's no depth limit on `directory_tree`, no size cap on `read_multiple_files` and no result cap on `search_files` (10). Errors name the problem and the fix, for example "Access denied - path outside allowed directories: X not in Y", "Destination already exists" and "Could not find exact match for edit", and `read_multiple_files` reports per-file failures without failing the batch (15). Every tool carries `readOnlyHint`. The four write tools set `destructiveHint` and `idempotentHint` accurately, and since 31 August `move_file` fails instead of overwriting (20). Mostly one required parameter (`path`). Node only, plus Docker (10). The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics): - 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries). - 20, pagination, filtering and output-size controls. - 20, actionable, documented error responses, codes and messages an agent can recover from. - 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations. - 15, sensible defaults, few required parameters, and official SDKs in at least two languages. Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs. ## 6. Maintenance & community, 60 out of 100, up to 3.5 more on the total Why it scored 60: 2026.8.31 on 31 August, 31 days before the run date (20). Three releases with a new filesystem version since 3 July, 2026.7.4, 2026.7.10 and 2026.8.31 (20). Fix pull requests for this server merged in July and August 2026, but #4702 has waited since 28 August with no maintainer reply and the repository has 221 open issues. GitHub's robots rules blocked issue search, so we read the first page only (12). Not in the official MCP registry. A lookup for io.github.modelcontextprotocol/server-filesystem returns 404 although package.json carries that `mcpName` (0). Dependabot, OIDC trusted publishing since July 2026, current SDK (^1.30.0) and zod 4, less the open Docker build failure (8). The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance): - 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older. - 20, at least three releases or dated changelog entries in the last 90 days. - 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15. - 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models). - 10, package health, current dependencies and CI. Models are read for deprecation notice periods and model churn rather than release counts. ## 7. Transparency & trust, 75 out of 100, up to 2.2 more on the total Made of editorial 76, provenance 74. Why it scored 75: MIT, inside a repository moving new contributions to Apache-2.0, both OSI licences (28). Local software that keeps nothing and calls no network service. There's no written statement, but the source is about 1,600 lines and shows it (20). `read_file` is marked deprecated in its own description with the replacement named, but there's no deprecation policy or dates (8). No telemetry in the code (20). The checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency): - 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms. - 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors). - 0 to 20, a deprecation policy or notices with dates. - 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted). The other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two. Provenance checks not met in full (half of this category, computed from checked facts): - Domain age: modelcontextprotocol.io, registered 2024-11-18 (1 year) (3 of 15) - Status page: not found (0 of 10) ## What we couldn't check What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it. - unchecked: whether the TypeScript CI workflow passes on the current head of main, since the Actions page we loaded showed older runs - unchecked: the full list of open filesystem issues, since GitHub's robots rules blocked issue search and we read the first page only - Whether the maintainers will deprecate the pre-2025.11.25 npm versions that break under zod v4 (#4702) ## Weaknesses - No depth limit on `directory_tree` and no size cap on reads or search results - No read-only mode in the server itself, only read-only Docker mounts - About 3,200 tokens of tool definitions by our estimate, with no toolsets to trim them - Not in the official MCP registry, and SECURITY.md declines vulnerability reports - The README still lists deleting directories as a feature, but no tool does that ## What costs an agent a turn today The notes we give agents before they call it. Each one is a workaround an agent shouldn't need. - Pass `excludePatterns` such as `node_modules` and `.git` to `directory_tree`. There's no depth parameter - Call `list_allowed_directories` first. Any path outside them fails - Use `head` or `tail` on `read_text_file` for logs and large files. Nothing else limits the size - Run `edit_file` with `dryRun: true` first. The edit fails if `oldText` doesn't match - Expect `move_file` to fail if the destination exists. Move or delete the target first ## What the review panel asked for - cut the filler from descriptions - fix the README directory-deletion claim - server-side read-only flag - accept vulnerability reports ## When it's done Send what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `"kind": "dispute"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.