# Augment Code (Augment)

Feedback Bench, coding agents, built 2026-10-19, window 2026-08-31 to 2026-10-18. Web page: https://feedbackbench.com/coding-agents/#/agent/augment

| Measure | Value |
|---|---|
| Rank | 17 of 17 (rank range 17–17) |
| Feedback Score | 34.1 (95% interval 33.6–34.6) |
| Popularity | 0.000 (share of voice 0.04%) |
| Customer love | 0.501 (95% interval 0.497–0.504) |
| Place | Popularity 17th of 17 × Customer love 11th of 17 |
| Top quadrant | no |
| Users | 67 |
| Posts counted | 81 |
| Posts that judge the agent | 35 |
| Criteria ranking high / low | 0 / 0 of 63 |

## The brief

Written by large language models from this page's numbers and the posts it quotes, then checked: every number in the text is in the build. Interpretation, not measurement.

Augment Code users praise how it handles instructions and context. Their complaints are about pricing and accounts. Instructions and context make up 68.2% of what its users rate, against 6.0% for other coding agents, and every rated week in that area is praise. Pricing and limits get the second-largest share, and nearly all of those weeks are complaints. Support and account issues get no praise. Few people talk about the agent at all. It ranks 17th of 17 overall and on popularity, and 11th on customer love. In this period, users did not compare it head to head with another agent.

### Why people choose it


### Why people leave


### What would move its score

Rule: For each area where an agent's customer love is below the category norm: the Feedback Score and rank it would have if that one area were at the norm (0.5) and everything else stayed the same, computed with the same formula as the ranking. An area's pull follows its weight, so an area many users of the category rate moves the score more than one they rarely mention.

| Area at the category norm | Feedback Score | Change | Position |
|---|---|---|---|
| Paying and limits | 34.2 | +0.1 | 17th of 17 |
| Doing the work | 34.2 | +0.1 | 17th of 17 |
| Choosing models | 34.1 | +0.0 | 17th of 17 |
| Account and support | 34.1 | +0.0 | 17th of 17 |

No single area moves the Feedback Score far. Bringing pricing and limits, or quality of work, up to the category norm raises the score by 0.1, to 34.2, and the agent stays 17th. Doing the same for models or for support and account leaves the score at 34.1.

### Six weeks

Weekly users ranged from 6 to 16. They peaked in the week of 2026-10-05, then fell back to 6. Share of voice stayed at 0% each week. The agent ranked 17th for five weeks, then 16th for the last two.

## Top requests

What users ask to add or change, most asked first. 7 user-weeks ask for something. Requests do not change the Feedback Score. Rule: A separate pass by a large language model reads every counted post and extracts what the user asks the agent or its vendor to add or change, with the criteria it maps to and a short normalised wording; it does not touch the labels or the Feedback Score. A large language model groups the wordings within each criterion (the first criterion the request maps to) into themes, and a second pass merges themes that ask for the same change under different criteria; a merged theme sits under the criterion most of its requests map to, and code counts it once. A theme counts distinct user-weeks that ask for it, per agent; across agents, one user-week per agent. Themes asked in fewer than 2 user-weeks, and requests that share no theme, are not shown. Examples: up to 3 posts per theme from different users, without slurs: whole posts of 60 to 450 characters with at least 8 words beyond handles and links, posts of at most 280 characters first, then the newest.

No request is asked for in enough user-weeks to show.

## Ranks highest on

No criterion where it sits clearly in the top half.

## Ranks lowest on

No criterion where it sits clearly in the bottom half.

Position on a criterion: its place among the agents with at least 30 rated user-weeks there, ordered by customer love. It ranks high when both its position and its 95% interval put it in the top half, low when both put it in the bottom half; neither otherwise, or when fewer than 4 agents qualify.

## All 63 criteria

Criterion love: 0.5 is the category norm. n: rated user-weeks.

### Paying and limits: Too few posts (customer love 0.496, n 10)

Users bring up pricing and limits more often than they do for other agents, and almost always to complain. Only one rated week is praise. Customer love is 0.496, close to the category norm. There are too few posts to rank it.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [How much use a plan's price buys](https://feedbackbench.com/coding-agents/criteria/limits.plan_value.md) | Too few posts | 0.498 | 0.494–0.502 | 5 | 1 | 4 |
| [Single prompt, model or effort level consumes disproportionate quota](https://feedbackbench.com/coding-agents/criteria/limits.burn_rate.md) | Too few posts | 0.495 | 0.491–0.499 | 5 | 0 | 5 |
| [Price, allowance or plan terms changed](https://feedbackbench.com/coding-agents/criteria/limits.allowance_change.md) | Too few posts | 0.499 | 0.497–0.500 | 2 | 0 | 2 |
| [Usage meter visibility and accuracy](https://feedbackbench.com/coding-agents/criteria/limits.usage_meter.md) | Too few posts | 0.498 | 0.496–0.500 | 2 | 0 | 2 |
| [Pricing and plan terms stated clearly and consistently](https://feedbackbench.com/coding-agents/criteria/billing.pricing_clarity.md) | Too few posts | 0.499 | 0.498–0.500 | 1 | 0 | 1 |
| [Free tier and free model availability and limits](https://feedbackbench.com/coding-agents/criteria/billing.free_tier.md) | Too few posts | 0.502 | 0.500–0.506 | 1 | 1 | 0 |
| [Short rolling usage window blocks or interrupts work](https://feedbackbench.com/coding-agents/criteria/limits.window_interrupts_work.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Quota reset timing and bonus or banked resets](https://feedbackbench.com/coding-agents/criteria/limits.reset_schedule.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Prompt cache hits, misses and invalidation](https://feedbackbench.com/coding-agents/criteria/limits.prompt_cache.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Pay-as-you-go overage, fallback billing and spend caps](https://feedbackbench.com/coding-agents/criteria/billing.overage_charges.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Using an existing subscription across tools](https://feedbackbench.com/coding-agents/criteria/billing.subscription_portability.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |

Most recent posts:

- Complaint, 2026-09-08, @augmentcode (X): “@augmentcode lol im using kimi 2.7 code and 22mins, it used up 12% of my plans credits. so a $20 sub wont even last 24hours?” [source](https://twitter.com/2002397531226148864/status/2097377115071287410)
- Complaint, 2026-09-18, @augmentcode (X): “@augmentcode regardless of whether it merges code or offers optimizations, the subscription for augment is very expensive. i ran out of tokens unexpectedly without even realizing it, and the pricing structure is unclear—i wouldn't recommend using it.” [source](https://twitter.com/955391612632252418/status/2100826713957810177)

### Setting up and connecting: Too few posts (customer love 0.505, n 8)

Users bring up setup 3.25 times as often as they do for other agents. Praise slightly outnumbers complaints. Customer love of 0.505 is close to the category norm.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [IDE and editor integration](https://feedbackbench.com/coding-agents/criteria/setup.ide_integration.md) | Too few posts | 0.504 | 0.500–0.508 | 5 | 4 | 1 |
| [Onboarding, discoverability and documentation](https://feedbackbench.com/coding-agents/criteria/setup.onboarding_docs.md) | Too few posts | 0.500 | 0.495–0.508 | 4 | 1 | 3 |
| [Install, launch and sign-in](https://feedbackbench.com/coding-agents/criteria/setup.install_signin.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Connecting own API keys, local models and custom endpoints](https://feedbackbench.com/coding-agents/criteria/setup.provider_byok_local.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [MCP servers, plugins, skills and hooks](https://feedbackbench.com/coding-agents/criteria/setup.extensions_mcp.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |

### Choosing models: Too few posts (customer love 0.499, n 1)

Users rarely bring up models. The only rated week is a complaint. Customer love of 0.499 is close to the category norm.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [Quality got worse or better over time](https://feedbackbench.com/coding-agents/criteria/models.quality_drift.md) | Too few posts | 0.499 | 0.498–0.500 | 1 | 0 | 1 |
| [Which models are offered on a plan and when](https://feedbackbench.com/coding-agents/criteria/models.catalog_access.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Automatic model routing and fallback](https://feedbackbench.com/coding-agents/criteria/models.routing_auto.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Reasoning effort setting and its defaults](https://feedbackbench.com/coding-agents/criteria/models.effort_control.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |

### Instructing and context: Too few posts (customer love 0.532, n 15)

Users discuss this area more than any other, and they praise it. It comes up 11.28 times as often as it does for other agents. Every rated week is praise. Customer love of 0.532 is the highest of any of its areas.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [Finding the right files in the codebase](https://feedbackbench.com/coding-agents/criteria/context.codebase_retrieval.md) | Too few posts | 0.527 | 0.516–0.539 | 15 | 15 | 0 |
| [Persistent project rules files are read and obeyed](https://feedbackbench.com/coding-agents/criteria/context.instruction_files.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Direct in-prompt instructions and caps are followed](https://feedbackbench.com/coding-agents/criteria/context.instruction_following.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Asks the user versus guessing](https://feedbackbench.com/coding-agents/criteria/context.clarifying_questions.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Output degrades as the context window fills](https://feedbackbench.com/coding-agents/criteria/context.long_context_decay.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Context compaction keeps what matters, cheaply and quickly](https://feedbackbench.com/coding-agents/criteria/context.compaction.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Memory and state carried across sessions](https://feedbackbench.com/coding-agents/criteria/context.session_memory.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Images, PDFs and file attachments as input](https://feedbackbench.com/coding-agents/criteria/context.attachments.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |

Most recent posts:

- Praise, 2026-10-08, @augmentcode (X): “@augmentcode @harnessio the context engine of augment was truely one of a kind, and i’m yet to this day to find anything as fast when it comes to context across large codebases. best of luck to the team!” [source](https://twitter.com/985631142043836417/status/2108245915488194661)
- Praise, 2026-10-08, @augmentcode (X): “@augmentcode @harnessio @igoro @scott_dietzen and team was early and built something special to be proud of. the context engine was innovative from day one, and delightful the first time i used it inside vscode. excited to see cosmos, auggie, and the context engine inside harness. onwards! [link]” [source](https://twitter.com/130935476/status/2108218530957341092)
- Praise, 2026-09-09, G2 (G2): “q: what problems is the product solving and how is that benefiting you? a: it supports me in code development and also has the ability to notify me of errors in extensive codes q: what do you like best about the product? a: the ability to understand the project and not just specific tasks q: what do you dislike about the product? a: as of today, there is nothing that displeases me as such” [source](https://www.g2.com/products/augment-code/reviews/augment-code-review-13431575)

### Doing the work: Too few posts (customer love 0.497, n 7)

Users bring up quality of work a little more often than they do for other agents, 31.8% against 26.8%. Praise and complaints are close to even. Customer love of 0.497 is close to the category norm.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [Can do the user's kind of task](https://feedbackbench.com/coding-agents/criteria/work.capability.md) | Too few posts | 0.499 | 0.492–0.505 | 5 | 3 | 2 |
| [Does unrequested work or over-engineers](https://feedbackbench.com/coding-agents/criteria/work.scope_overreach.md) | Too few posts | 0.498 | 0.495–0.500 | 2 | 0 | 2 |
| [Frontend and visual UI output](https://feedbackbench.com/coding-agents/criteria/work.frontend_ui.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Diagnosing and fixing reported bugs](https://feedbackbench.com/coding-agents/criteria/work.bug_diagnosis.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Breaks existing code or reintroduces bugs](https://feedbackbench.com/coding-agents/criteria/work.regressions_introduced.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Spins, loops or gets stuck without progress](https://feedbackbench.com/coding-agents/criteria/work.stuck_loops.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Stops mid-task or answers instead of acting](https://feedbackbench.com/coding-agents/criteria/work.premature_stop.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Long unattended runs and goal/loop mode](https://feedbackbench.com/coding-agents/criteria/work.long_running_autonomy.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Subagents, parallel agents and orchestrators](https://feedbackbench.com/coding-agents/criteria/work.multi_agent_orchestration.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Games checks instead of fixing the problem](https://feedbackbench.com/coding-agents/criteria/work.reward_hacking.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Risky or irreversible actions without confirmation](https://feedbackbench.com/coding-agents/criteria/work.destructive_actions.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Git commits, branches and sync](https://feedbackbench.com/coding-agents/criteria/work.git_workflow.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Computer use and browser control](https://feedbackbench.com/coding-agents/criteria/work.computer_browser_use.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Safety filters block legitimate coding tasks](https://feedbackbench.com/coding-agents/criteria/work.safety_refusals.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Tool approval prompts and autonomy modes](https://feedbackbench.com/coding-agents/criteria/work.permission_prompts.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Plan-before-edit mode](https://feedbackbench.com/coding-agents/criteria/work.plan_mode.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Length and clarity of replies, summaries and comments](https://feedbackbench.com/coding-agents/criteria/work.response_verbosity.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Caves to or argues with the user's judgement](https://feedbackbench.com/coding-agents/criteria/work.sycophancy_pushback.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |

### Checking and finishing: Too few posts (customer love 0.501, n 1)

Users rarely bring up checking the work. The one rated week is praise. Customer love of 0.501 is close to the category norm.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [Agent-performed code review finds real issues](https://feedbackbench.com/coding-agents/criteria/verify.agent_code_review.md) | Too few posts | 0.501 | 0.500–0.503 | 1 | 1 | 0 |
| [Claims work is done or fixed when it is not](https://feedbackbench.com/coding-agents/criteria/verify.false_completion.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Builds, tests or runs its own changes](https://feedbackbench.com/coding-agents/criteria/verify.self_testing.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Reviewing and approving the agent's changes](https://feedbackbench.com/coding-agents/criteria/verify.change_review_ui.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |

### Interface and sessions: Too few posts (customer love 0.500, n 2)

Users bring up the interface less often than they do for other agents. Praise and complaints are even. Customer love sits at the category norm.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [Mobile, remote-control and voice access](https://feedbackbench.com/coding-agents/criteria/surfaces.remote_mobile.md) | Too few posts | 0.501 | 0.500–0.502 | 1 | 1 | 0 |
| [Cloud and remote sandbox execution](https://feedbackbench.com/coding-agents/criteria/surfaces.cloud_sessions.md) | Too few posts | 0.499 | 0.496–0.500 | 1 | 0 | 1 |
| [How the interface shows work, and what the user can configure](https://feedbackbench.com/coding-agents/criteria/ui.display_settings.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Saving, switching, resuming and rewinding sessions](https://feedbackbench.com/coding-agents/criteria/ui.session_history.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Stopping and steering a running agent](https://feedbackbench.com/coding-agents/criteria/ui.interrupt_steer.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |

### Reliability and speed: Too few posts (customer love 0.504, n 3)

Reliability comes up about as often as it does for other agents, 13.6% against 13.2%. Complaints slightly outnumber praise. Customer love of 0.504 is close to the category norm.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [Latency, throughput and fast mode](https://feedbackbench.com/coding-agents/criteria/rel.response_speed.md) | Too few posts | 0.503 | 0.496–0.512 | 2 | 1 | 1 |
| [Updates break working setups](https://feedbackbench.com/coding-agents/criteria/rel.update_breakage.md) | Too few posts | 0.499 | 0.497–0.500 | 1 | 0 | 1 |
| [Outages, server errors and capacity or rate errors](https://feedbackbench.com/coding-agents/criteria/rel.service_errors.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Client crashes, freezes and failed tool execution](https://feedbackbench.com/coding-agents/criteria/rel.client_failures.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |

### Account and support: Too few posts (customer love 0.497, n 4)

Users bring up support and account issues 3.21 times as often as they do for other agents. Every rated week is a complaint. Customer love is 0.497, just under the category norm.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [Support, refunds and issue handling](https://feedbackbench.com/coding-agents/criteria/account.support.md) | Too few posts | 0.497 | 0.494–0.499 | 4 | 0 | 4 |
| [Wrong charges, failed payments and plan provisioning](https://feedbackbench.com/coding-agents/criteria/account.billing_errors.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Account bans and access restrictions](https://feedbackbench.com/coding-agents/criteria/account.bans_restrictions.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Data retention, training use and deployment isolation](https://feedbackbench.com/coding-agents/criteria/account.data_privacy.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |