Auggie CLI

# Fix list: Auggie CLI

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/auggie, the October 2026 research run, assessed 18 October 2026. Grade D, 50.8 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 Auggie CLI: 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. Reliability, 43 out of 100, up to 11.4 more on the total

Why it scored 43: Local-package reading, as for the other closed-source harnesses. The official npm package with macOS, Linux and WSL stated, but the README asks for Node 22 or later while the docs and the package engines say Node 20, so 2 are held back (18). No public CI or test suite, since the source isn't published. The repository has six workflow files, but it holds docs, examples and issues, not the CLI, and their results couldn't be read (0). The repository's tracker had 93 open and 18 closed issues on 18 October 2026, per our GitHub record, so most reports stay open. Which are crashes or regressions, and how fast maintainers reply, couldn't be read. Every run needs Augment's servers, and the status page records major outages on 1 September 2026 (Cosmos and the CLI unavailable, about 40 minutes) and 9 October 2026 (about five and a half hours across services) (5). A changelog by version with fixes and improvements in sections, dated release notes on the website, and a 1.x guide with a breaking-changes section, though the repository changelog carries no dates (10). The `latest` 0.x channel is described as the current stable line for production, while 1.x is in preview. Judgement call (10). 18+0+5+10+10 = 43.

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.

## 2. Payments & pricing, 20 out of 100, up to 10 more on the total

Why it scored 20: Harness reading of the published rubric, scored on the Augment subscription the CLI needs. The self-hosted rule doesn't apply. No payment protocol (0). Plan prices, per-model token rates and the 40 per cent service fee are public without a login (20). No free tier was found on the pricing page. The terms mention trials by order (0). A person signs in through a browser, and service account tokens are created by an administrator in the web app (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.

## 3. Security & auth, 47 out of 100, up to 9.3 more on the total

Why it scored 47: Framework reading (telemetry defaults, approvals, guardrails, sandboxing), five lines as for the other harnesses. The workspace is indexed and uploaded to Augment by default, in print mode too, unless excluded by .gitignore or .augmentignore. A telemetry endpoint is listed with no description or opt-out found. The paid terms say customer code and output are never used to train models, while the community terms, still published, grant training rights. Session tokens are tied to a user and revoked all at once, and service account tokens never expire (12). Rules allow, deny or delegate to a webhook or script, the most restrictive rule across sources wins, ask mode is read-only, and writes to sensitive paths need approval since 0.34.0. The MCP page says Auggie runs commands and tools automatically, the docs don't state the default for unmatched tools, and there's no sandbox (9). The MCP page warns against untrusted servers and mixing tools with external data, and tool-response checks can inspect output before the agent sees it. No network control was found (6). Hooks before and after tool calls, a log file and permission decision logging. Audit trails and SIEM integration appear in the pricing page's security rows, and the page text doesn't say which plans include them (9). A disclosure address and GitHub private reporting in SECURITY.md, SOC 2 Type II and ISO/IEC 42001 named. No security.txt and no bug bounty found. OSV lists no published advisory for the package and NVD no CVE for Auggie, per our GitHub record, so there's no public advisory record to credit or deduct (11). 12+9+6+9+11 = 47.

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.

## 4. Agent ergonomics, 65 out of 100, up to 5.7 more on the total

Why it scored 65: Framework reading, adapted to a harness driven by a pipeline. `--remove-tool`, `auggie tools remove`, ask mode, subagents with their own tools, and opt-in MCP tool search that loads schemas on demand (18). `--max-turns`, `--quiet`, `--compact`, `--output-format json`, `--queue` and `--show-credits`. Budgets per user exist only on Enterprise (15). No documented exit codes or error format for print mode. Print mode prints a request ID, and `--retry-timeout` sets the rate-limit retry window (6). `--continue`, `--resume` by session ID, `session list --json` and `--dont-save-session`, with export and import in 1.x (15). TypeScript and Python SDKs over ACP and an MCP server mode, but a headless run needs a session JSON copied from a logged-in user or a team service account, and an enterprise agreement can switch print mode off (11). 18+15+6+15+11 = 65.

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.

## 5. Schema & documentation, 67 out of 100, up to 5.4 more on the total

Why it scored 67: Framework reading. `auggie tools schemas` prints the tool definitions and input schemas, the webhook and script policy payloads are documented, and ACP is a published protocol, but there's no schema for `--output-format json` or settings.json (13). docs.augmentcode.com/llms.txt indexes every page with a Markdown twin (10). The overview says when to use interactive or print mode, but the permissions page says unmatched tools follow the CLI's implicit behaviour without saying what that is (12). Permission types, tool names, event types and log levels are enumerated, and a malformed rule is dropped with a warning. The 1.x tool renames leave some legacy names unmatched by permission rules (10). Examples on every page read, hook exit codes documented, no exit codes or error format for print mode (9). Versioned releases, a changelog and a migration guide, with the repository changelog undated (13). 13+10+12+10+9+13 = 67.

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.

## 6. Transparency & trust, 57 out of 100, up to 3.8 more on the total

Made of editorial 35, provenance 79.

Why it scored 57: Proprietary. The licence is clear on subscription, redistribution and modification, but it says commercial use needs a separate agreement while plans are sold to businesses, and it names Augment Computing, Inc., which sold the CLI to Harness on 8 October (12). The paid terms rule out training on customer code and treat it as confidential. The privacy policy (28 August 2025) gives no retention periods and lists no subprocessors, referring both to a request, and the docs say indexed code is stored by Augment without a retention period (12). The 1.x guide lists breaking changes and retired parts, and the terms promise no materially reduced functionality during an order, but no date for 1.x becoming the default and no notice period were found (8). A telemetry endpoint is disclosed on the network page, with no description of what's sent and no opt-out (3). 12+12+8+3 = 35.

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: augmentcode.com, registered 2022-06-16 (4 years) (7 of 15)
- security.txt: not found (0 of 10)

## 7. Maintenance & community, 58 out of 100, up to 3.7 more on the total

Why it scored 58: Graded on the stable line, as the brief asks. 0.36.0 on 21 August 2026, 58 days before this check (20). The 1.x preview and the 0.37.0 prerelease line had builds on 16 October, with 283 prereleases since 20 July. Judgement call to score the stable line. 0.34.0, 0.35.0 and 0.36.0 between 30 July and 21 August 2026 (20). The public tracker had 93 open and 18 closed issues, and 66 closed and no open pull requests, on 18 October 2026, per our GitHub record. Reply times couldn't be read. A support portal, a subreddit and a dated website changelog (6). TypeScript SDK 0.2.0 of 8 July 2026, and its docs still name older models. The Python SDK's PyPI page answered with a bot check (8). npm builds come from GitHub Actions through trusted publishing with registry signatures, and the changelog records dependency scan fixes. No public CI for the CLI. The repository's six workflow files don't build it, and their results couldn't be read. Its newest GitHub release is v0.28.0 from May 2026, while tags follow npm builds up to 16 October (4). 20+20+6+8+4 = 58.

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.

## 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: CI results, maintainer response times and the star count for augmentcode/auggie, which the GitHub data file couldn't read
- unchecked: the Python SDK's version and date, because the PyPI project page answered with a bot check
- unchecked: the trust centre at trust.augmentcode.com, so subprocessors and certification reports weren't read
- What the CLI does by default with a tool no permission rule matches, in interactive and in print mode
- What the telemetry endpoint receives and whether it can be turned off
- How long indexed code is kept on Augment's servers
- Whether Harness will move the package, docs, terms or contracting entity, and when 1.x becomes the `latest` channel
- Whether the community terms, which allow training on code, still apply to any current plan

## Weaknesses

- No operating-system sandbox, and the MCP docs say Auggie runs commands and tools automatically
- Print mode indexes and uploads the workspace by default, except files matched by .gitignore or .augmentignore
- No free tier on the pricing page, and print mode may be switched off under an enterprise agreement
- The stable npm line is 0.36.0 of 21 August 2026, with 1.x builds only on the preview and daemon channels
- The licence, terms and privacy policy still name Augment Computing, Inc. after Harness bought the CLI on 8 October 2026

## 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 `--max-turns` with `--print` in pipelines to bound how many turns, and so how much usage, a run can take
- Commit deny rules to .augment/settings.json before an unattended run. Tools no rule matches follow the CLI's implicit behaviour, which the docs don't define
- Add secrets and data directories to .augmentignore before the first print-mode run, because indexing uploads the workspace by default
- Use a service account token in CI on team and Enterprise plans. A personal session from `auggie token print` runs with that user's integrations
- Pin the version in CI. `latest` is 0.x, while `preview` and `daemon` carry 1.x prereleases whose permission rules ignore some legacy tool names

## What the review panel asked for

- date for 1.x promotion
- dated repository changelog
- per-user budgets below Enterprise
- JSON output schema
- state the unmatched-tool default
- document unmatched-tool default
- expiring service account tokens

## 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.