{
  "fixes": {
    "slug": "neckbeard",
    "name": "neckbeard",
    "listing": "https://www.anchorterminal.com/tools/neckbeard",
    "markdown": "# Fix list: neckbeard\n\nFrom Anchor Terminal's listing at https://www.anchorterminal.com/tools/neckbeard, the October 2026 research run, assessed 17 October 2026. Grade B, 64.2 out of 100.\n\nThis 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.\n\nFor a coding agent working on neckbeard: 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.\n\n## 1. Reliability, 53 out of 100, up to 9.4 more on the total\n\nWhy it scored 53: Read with the local-software lines, since the owner runs the plugin and the CLI. The project is about six weeks old (first commit 8 September 2026) and was launched on Hacker News on 14 October. The Go module is on pkg.go.dev and the plugin's installer fetches a checksummed release binary for Linux or macOS, falling back to `go install`. The Go version is stated only in go.mod (1.27.1), and the release assets themselves couldn't be seen (17 of 20). A public test workflow runs gofmt, `go vet`, `go test`, the plugin and installer checks and a six-lane plan, scaffold and validate matrix on every pull request and push to main. We couldn't see whether main passes (20 of 25). GitHub's issue pages couldn't be read, so there is no readable record of open bugs. INSTALL.md links one open issue, #27 (10 of 25). Semantic version tags and a written semver contract for the catalogue, but no changelog file, and the plugin-first change in 0.2.0 was not marked as breaking (6 of 15). The README calls it pre-alpha at 0.3.0 (0). Total 53.\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):\n\nHosted APIs, MCP servers, models and platforms.\n\n- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).\n- 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.\n- 15, rate limits documented with numbers.\n- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.\n- 10, an SLA published for any paid tier.\n- 10, the surface agents use is generally available, not beta or preview.\n\nLocal packages, SDKs, frameworks and stdio MCP servers.\n\n- 20, installs from an official package with supported runtimes stated.\n- 25, a public CI and test suite, passing on the default branch.\n- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).\n- 15, semver discipline and breaking changes called out in a changelog.\n- 15, version 1.0 or later, or declared stable.\n\nProtocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.\n\n## 2. Security \u0026 auth, 68 out of 100, up to 5.6 more on the total\n\nWhy it scored 68: Read with the framework lines, as for other local agent tools, since neckbeard holds no secret of its own. Analysis, planning, scaffolding and validation need no cloud credentials, and the skill and README say the agent never gets them. Generated pipelines use OIDC federation with no static keys, and a test fails if a rendered pipeline mentions `AWS_SECRET_ACCESS_KEY`. No telemetry found in the source. `estimate` relies on Infracost's Cloud Pricing API, and the experimental `release` commands can run with operator credentials from a workstation (25 of 30). Generated pipelines plan under a read-only identity and apply on main under a write identity, with production behind environment protection. `foundation-apply` refuses updates, replacements and deletes after a person reviews the saved plan (17 of 20). The agent reads the user's own repository. No guidance on untrusted content in it, beyond never transcribing secret values (8 of 15). The blueprint is a hash-verified lockfile, release receipts are committed to the repository, and validation reports each check. No log of agent actions (9 of 15). SECURITY.md with GitHub private vulnerability reporting and a stated scope, no staffed response time, no bug bounty, and advisories unread. Actions pinned by commit SHA, a fork-safety CI job and a checksum-checked installer (9 of 20). Total 68.\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-security):\n\n- 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.\n- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.\n- 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.\n- 0 to 15, audit logs or per-call visibility for the operator.\n- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.\n\nModels 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.\n\n## 3. Payments \u0026 pricing, 60 out of 100, up to 5 more on the total\n\nWhy it scored 60: No payment protocol (0). Read with the self-hosted rule. Free under Apache-2.0 with nothing to buy (20). No card or account for the plugin (20). Installs from the plugin marketplace or `go install` with no signup (20). Estimates need a person to sign in to Infracost, which is outside the product and not deducted. Total 60.\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):\n\nThe published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).\n\n- 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.\n- 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.\n- 20, a free tier or trial that doesn't need a card.\n- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).\n\nPayment 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.\n\nOpen-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.\n\n## 4. Agent ergonomics, 71 out of 100, up to 4.7 more on the total\n\nWhy it scored 71: No tool definitions. INSTALL.md puts the always-on SessionStart context at about 280 tokens, and SKILL.md (about 9 KB) and two references (about 13 KB) load when needed. Results go to files on disk with a short summary on stdout (20 of 25). `doctor -for` narrows checks and `estimate` takes a comparison argument, with nothing to page (10 of 20). Errors name the fix, such as the ceiling error that lists four ways to bring the total down. Exit codes are 0, 1 and 2, with no code per failure class (14 of 20). Same inputs render the same bytes, regeneration refuses to overwrite edited files and writes a `.neckbeard-new` copy, `analyze` won't overwrite a profile without `-force`, and `prepare-database` stores its password first so a retry reuses it (18 of 20). Defaults for every path flag and tier presets. One CLI and no library in a second language (9 of 15). Total 71.\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):\n\n- 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).\n- 20, pagination, filtering and output-size controls.\n- 20, actionable, documented error responses, codes and messages an agent can recover from.\n- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.\n- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.\n\nModels 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.\n\n## 5. Transparency \u0026 trust, 52 out of 100, up to 4.2 more on the total\n\nMade of editorial 63, provenance 40.\n\nWhy it scored 52: Apache-2.0 in LICENSE and plugin.json (30). No privacy statement. The README says secret values never pass through neckbeard, its state or git history, and the source we read has no telemetry. What `estimate` sends to Infracost's pricing service is not described (15 of 30). DESIGN.md section 6.3 commits the catalogue to stable input schemas and resource addresses within a major version, with a migration path for a major, and allows planner default changes before 1.0 with migration notes. No dated deprecation notices (8 of 20). No telemetry found, and the docs don't address it (10 of 20). Total 63.\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):\n\n- 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.\n- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).\n- 0 to 20, a deprecation policy or notices with dates.\n- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).\n\nThe other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.\n\nProvenance checks not met in full (half of this category, computed from checked facts):\n\n- Domain age: , no registry record we could read (0 of 15)\n- Status page: not found (0 of 10)\n- Changelog: not found (0 of 10)\n- security.txt: not found (0 of 10)\n\n## 6. Schema \u0026 documentation, 75 out of 100, up to 4.1 more on the total\n\nWhy it scored 75: Graded on the CLI and its skill. JSON Schema 2020-12 files for `neckbeard.yaml`, `app-profile.yaml`, the blueprint and the environment manifest, printed by `neckbeard schema`. Command output is text, with the estimate also written as JSON (18 of 25). The skill, two reference files and five commands are Markdown written for agents (10). The skill's description says when to load it, and SKILL.md says what the agent must not do, such as hand-write HCL, map an unsupported need onto the nearest module or read secret values (17 of 20). The schemas use enums, required fields and `additionalProperties: false`, and `neckbeard presets` prints the allowed tiers (13 of 15). The onboarding reference has a complete config and examples for existing databases and unsupported needs. Errors are plain `error:` lines with no catalogue, though many name the next step (9 of 15). Semantic version tags on the CLI and the catalogue, and generated GitHub release notes we couldn't read, with no changelog file (8 of 15). Total 75.\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):\n\nAPIs and MCP servers.\n\n- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).\n- 10, llms.txt or Markdown docs served for agents.\n- 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.\n- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.\n- 0 to 15, examples and documented error responses.\n- 15, versioning and a public changelog.\n\nModels 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.\n\n## 7. Maintenance \u0026 community, 68 out of 100, up to 2.8 more on the total\n\nWhy it scored 68: v0.3.0 on 15 October 2026 (30). v0.1.0 (commit of 11 September), v0.2.0 and v0.2.1 (14 October) and v0.3.0 (15 October) inside 90 days, plus catalogue tags (20). GitHub's issue and pull request pages couldn't be read. The author answered about 20 comments on the Hacker News thread on 14 and 15 October, and two pull requests from another account were merged on 12 September and 15 October (10 of 25). Not an MCP server and no SDK, and the plugin is in its own marketplace, not an official one (0). Dependabot runs weekly on actions and Go modules, actions are pinned by SHA and CI runs on every push. CI status unread (8 of 10). Total 68.\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):\n\n- 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.\n- 20, at least three releases or dated changelog entries in the last 90 days.\n- 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.\n- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).\n- 10, package health, current dependencies and CI.\n\nModels are read for deprecation notice periods and model churn rather than release counts.\n\n## What we couldn't check\n\nWhat 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.\n\n- Judgement call. Listed in infrastructure with `infra.terraform` first, as the scout suggested, because its output is OpenTofu modules and pipelines. `code.deploy` is left out because the `release` workflow is experimental and its live passes are the project's own claim.\n- `firstReleased` is 2026-09-11, the date of the commit tag v0.1.0 points at, which pkg.go.dev also lists. The tag is lightweight, so the day it was pushed and the day the repository went public are not known. v0.2.0 on 14 October 2026, the plugin-first release, and the Show HN post the same day are the launch.\n- unchecked: GitHub web pages and API (stars, issues, pull request timelines, CI run status, releases and their assets, security advisories, the author's profile). The proxy refused them, so the repository was read by git clone.\n- Lapse. proxy.golang.org was fetched in the same command as its robots.txt, which disallows those paths. The responses were discarded and nothing from them is used; the version dates come from pkg.go.dev and the git tags.\n- What `estimate` sends to Infracost's Cloud Pricing API was not checked. Infracost is a separate product.\n- Neither ours. It doesn't do the job of LocalGhost or letme. We did not install or run the tool, and every claim about behaviour comes from the source and the project's documents.\n\n## Weaknesses\n\n- Pre-alpha at version 0.3.0, and the README marks GCP, GitLab, Kubernetes and promotion as not exercised live\n- No changelog file. Release notes are generated by GitHub, and the move to a plugin layout in 0.2.0 was not marked as breaking\n- Estimates need the older Infracost 0.10.x CLI signed in to Infracost, and the tool refuses the current Homebrew formula\n- CLI output is plain text with exit codes 0, 1 and 2 only. No JSON output flag\n- One maintainer, with SECURITY.md saying there is no staffed response time\n\n## What costs an agent a turn today\n\nThe notes we give agents before they call it. Each one is a workaround an agent shouldn't need.\n\n- Run `neckbeard doctor -for plan` first, and `-for estimate` before estimating, to see which tools are missing\n- Answer every assumption in `app-profile.yaml` with a `confirmed` entry of the same id, since `plan` refuses while any stays open\n- Never edit `blueprint.yaml`. Change the profile or `neckbeard.yaml` and run `neckbeard plan` again\n- Show the person `costs/estimate.md` and its unpriced section before running `scaffold`\n- Hand bootstrap and any `release` step to the person holding cloud credentials. Never ask for secret values\n\n## What the review panel asked for\n\n- a changelog file\n- Document what estimate sends to Infracost\n- Add a JSON output flag\n- Document the error classes\n- staffed disclosure response\n\n## When it's done\n\nSend 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.\n",
    "grade": "B",
    "score": 64.2,
    "assessed": "2026-10-17",
    "run": "October 2026 research run",
    "categories": [
      {
        "key": "reliability",
        "name": "Reliability",
        "score": 53,
        "maxGain": 9.4,
        "reason": "Read with the local-software lines, since the owner runs the plugin and the CLI. The project is about six weeks old (first commit 8 September 2026) and was launched on Hacker News on 14 October. The Go module is on pkg.go.dev and the plugin's installer fetches a checksummed release binary for Linux or macOS, falling back to `go install`. The Go version is stated only in go.mod (1.27.1), and the release assets themselves couldn't be seen (17 of 20). A public test workflow runs gofmt, `go vet`, `go test`, the plugin and installer checks and a six-lane plan, scaffold and validate matrix on every pull request and push to main. We couldn't see whether main passes (20 of 25). GitHub's issue pages couldn't be read, so there is no readable record of open bugs. INSTALL.md links one open issue, #27 (10 of 25). Semantic version tags and a written semver contract for the catalogue, but no changelog file, and the plugin-first change in 0.2.0 was not marked as breaking (6 of 15). The README calls it pre-alpha at 0.3.0 (0). Total 53.",
        "checklist": [
          "Hosted APIs, MCP servers, models and platforms.",
          "- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).\n- 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.\n- 15, rate limits documented with numbers.\n- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.\n- 10, an SLA published for any paid tier.\n- 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.\n- 25, a public CI and test suite, passing on the default branch.\n- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).\n- 15, semver discipline and breaking changes called out in a changelog.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-reliability"
      },
      {
        "key": "security",
        "name": "Security \u0026 auth",
        "score": 68,
        "maxGain": 5.6,
        "reason": "Read with the framework lines, as for other local agent tools, since neckbeard holds no secret of its own. Analysis, planning, scaffolding and validation need no cloud credentials, and the skill and README say the agent never gets them. Generated pipelines use OIDC federation with no static keys, and a test fails if a rendered pipeline mentions `AWS_SECRET_ACCESS_KEY`. No telemetry found in the source. `estimate` relies on Infracost's Cloud Pricing API, and the experimental `release` commands can run with operator credentials from a workstation (25 of 30). Generated pipelines plan under a read-only identity and apply on main under a write identity, with production behind environment protection. `foundation-apply` refuses updates, replacements and deletes after a person reviews the saved plan (17 of 20). The agent reads the user's own repository. No guidance on untrusted content in it, beyond never transcribing secret values (8 of 15). The blueprint is a hash-verified lockfile, release receipts are committed to the repository, and validation reports each check. No log of agent actions (9 of 15). SECURITY.md with GitHub private vulnerability reporting and a stated scope, no staffed response time, no bug bounty, and advisories unread. Actions pinned by commit SHA, a fork-safety CI job and a checksum-checked installer (9 of 20). Total 68.",
        "checklist": [
          "- 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.\n- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.\n- 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.\n- 0 to 15, audit logs or per-call visibility for the operator.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-security"
      },
      {
        "key": "payments",
        "name": "Payments \u0026 pricing",
        "score": 60,
        "maxGain": 5,
        "reason": "No payment protocol (0). Read with the self-hosted rule. Free under Apache-2.0 with nothing to buy (20). No card or account for the plugin (20). Installs from the plugin marketplace or `go install` with no signup (20). Estimates need a person to sign in to Infracost, which is outside the product and not deducted. Total 60.",
        "checklist": [
          "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.\n- 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.\n- 20, a free tier or trial that doesn't need a card.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-payments"
      },
      {
        "key": "ergonomics",
        "name": "Agent ergonomics",
        "score": 71,
        "maxGain": 4.7,
        "reason": "No tool definitions. INSTALL.md puts the always-on SessionStart context at about 280 tokens, and SKILL.md (about 9 KB) and two references (about 13 KB) load when needed. Results go to files on disk with a short summary on stdout (20 of 25). `doctor -for` narrows checks and `estimate` takes a comparison argument, with nothing to page (10 of 20). Errors name the fix, such as the ceiling error that lists four ways to bring the total down. Exit codes are 0, 1 and 2, with no code per failure class (14 of 20). Same inputs render the same bytes, regeneration refuses to overwrite edited files and writes a `.neckbeard-new` copy, `analyze` won't overwrite a profile without `-force`, and `prepare-database` stores its password first so a retry reuses it (18 of 20). Defaults for every path flag and tier presets. One CLI and no library in a second language (9 of 15). Total 71.",
        "checklist": [
          "- 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).\n- 20, pagination, filtering and output-size controls.\n- 20, actionable, documented error responses, codes and messages an agent can recover from.\n- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-ergonomics"
      },
      {
        "key": "transparency",
        "name": "Transparency \u0026 trust",
        "score": 52,
        "maxGain": 4.2,
        "reason": "Apache-2.0 in LICENSE and plugin.json (30). No privacy statement. The README says secret values never pass through neckbeard, its state or git history, and the source we read has no telemetry. What `estimate` sends to Infracost's pricing service is not described (15 of 30). DESIGN.md section 6.3 commits the catalogue to stable input schemas and resource addresses within a major version, with a migration path for a major, and allows planner default changes before 1.0 with migration notes. No dated deprecation notices (8 of 20). No telemetry found, and the docs don't address it (10 of 20). Total 63.",
        "blend": "editorial 63, provenance 40",
        "checklist": [
          "- 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.\n- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).\n- 0 to 20, a deprecation policy or notices with dates.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-transparency"
      },
      {
        "key": "schema",
        "name": "Schema \u0026 documentation",
        "score": 75,
        "maxGain": 4.1,
        "reason": "Graded on the CLI and its skill. JSON Schema 2020-12 files for `neckbeard.yaml`, `app-profile.yaml`, the blueprint and the environment manifest, printed by `neckbeard schema`. Command output is text, with the estimate also written as JSON (18 of 25). The skill, two reference files and five commands are Markdown written for agents (10). The skill's description says when to load it, and SKILL.md says what the agent must not do, such as hand-write HCL, map an unsupported need onto the nearest module or read secret values (17 of 20). The schemas use enums, required fields and `additionalProperties: false`, and `neckbeard presets` prints the allowed tiers (13 of 15). The onboarding reference has a complete config and examples for existing databases and unsupported needs. Errors are plain `error:` lines with no catalogue, though many name the next step (9 of 15). Semantic version tags on the CLI and the catalogue, and generated GitHub release notes we couldn't read, with no changelog file (8 of 15). Total 75.",
        "checklist": [
          "APIs and MCP servers.",
          "- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).\n- 10, llms.txt or Markdown docs served for agents.\n- 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.\n- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.\n- 0 to 15, examples and documented error responses.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-schema"
      },
      {
        "key": "maintenance",
        "name": "Maintenance \u0026 community",
        "score": 68,
        "maxGain": 2.8,
        "reason": "v0.3.0 on 15 October 2026 (30). v0.1.0 (commit of 11 September), v0.2.0 and v0.2.1 (14 October) and v0.3.0 (15 October) inside 90 days, plus catalogue tags (20). GitHub's issue and pull request pages couldn't be read. The author answered about 20 comments on the Hacker News thread on 14 and 15 October, and two pull requests from another account were merged on 12 September and 15 October (10 of 25). Not an MCP server and no SDK, and the plugin is in its own marketplace, not an official one (0). Dependabot runs weekly on actions and Go modules, actions are pinned by SHA and CI runs on every push. CI status unread (8 of 10). Total 68.",
        "checklist": [
          "- 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.\n- 20, at least three releases or dated changelog entries in the last 90 days.\n- 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.\n- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).\n- 10, package health, current dependencies and CI.",
          "Models are read for deprecation notice periods and model churn rather than release counts."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-maintenance"
      }
    ],
    "provenance": [
      {
        "label": "Domain age",
        "value": ", no registry record we could read",
        "points": 0,
        "max": 15
      },
      {
        "label": "Status page",
        "value": "not found",
        "points": 0,
        "max": 10
      },
      {
        "label": "Changelog",
        "value": "not found",
        "points": 0,
        "max": 10
      },
      {
        "label": "security.txt",
        "value": "not found",
        "points": 0,
        "max": 10
      }
    ],
    "unchecked": [
      "Judgement call. Listed in infrastructure with `infra.terraform` first, as the scout suggested, because its output is OpenTofu modules and pipelines. `code.deploy` is left out because the `release` workflow is experimental and its live passes are the project's own claim.",
      "`firstReleased` is 2026-09-11, the date of the commit tag v0.1.0 points at, which pkg.go.dev also lists. The tag is lightweight, so the day it was pushed and the day the repository went public are not known. v0.2.0 on 14 October 2026, the plugin-first release, and the Show HN post the same day are the launch.",
      "unchecked: GitHub web pages and API (stars, issues, pull request timelines, CI run status, releases and their assets, security advisories, the author's profile). The proxy refused them, so the repository was read by git clone.",
      "Lapse. proxy.golang.org was fetched in the same command as its robots.txt, which disallows those paths. The responses were discarded and nothing from them is used; the version dates come from pkg.go.dev and the git tags.",
      "What `estimate` sends to Infracost's Cloud Pricing API was not checked. Infracost is a separate product.",
      "Neither ours. It doesn't do the job of LocalGhost or letme. We did not install or run the tool, and every claim about behaviour comes from the source and the project's documents."
    ],
    "weaknesses": [
      "Pre-alpha at version 0.3.0, and the README marks GCP, GitLab, Kubernetes and promotion as not exercised live",
      "No changelog file. Release notes are generated by GitHub, and the move to a plugin layout in 0.2.0 was not marked as breaking",
      "Estimates need the older Infracost 0.10.x CLI signed in to Infracost, and the tool refuses the current Homebrew formula",
      "CLI output is plain text with exit codes 0, 1 and 2 only. No JSON output flag",
      "One maintainer, with SECURITY.md saying there is no staffed response time"
    ],
    "agentNotes": [
      "Run `neckbeard doctor -for plan` first, and `-for estimate` before estimating, to see which tools are missing",
      "Answer every assumption in `app-profile.yaml` with a `confirmed` entry of the same id, since `plan` refuses while any stays open",
      "Never edit `blueprint.yaml`. Change the profile or `neckbeard.yaml` and run `neckbeard plan` again",
      "Show the person `costs/estimate.md` and its unpriced section before running `scaffold`",
      "Hand bootstrap and any `release` step to the person holding cloud credentials. Never ask for secret values"
    ],
    "requests": [
      {
        "text": "a changelog file",
        "reviews": 1
      },
      {
        "text": "Document what estimate sends to Infracost",
        "reviews": 1
      },
      {
        "text": "Add a JSON output flag",
        "reviews": 1
      },
      {
        "text": "Document the error classes",
        "reviews": 1
      },
      {
        "text": "staffed disclosure response",
        "reviews": 1
      }
    ],
    "recheck": "https://www.anchorterminal.com/builders/#disputes"
  },
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "attributionRequired": true,
    "credit": "CC BY 4.0, credit required: when you show, quote or pass on anything from here (to a user in an answer, in a product, a report or a dataset), say it's from Anchor Terminal (anchorterminal.com) and link the page or API address you took it from. Reviews also credit their author.",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-10-18",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.4",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run",
    "terms": "https://www.anchorterminal.com/terms/#data"
  }
}