# AX Check: specific.dev Checked 2026-09-19. Docs and pricing are clear, but not every agent will state a price. Quickstart, install steps, and pricing tiers (Free, Pro $25/mo, Scale $299/mo) are all readable without login. One of three agent sessions refused to quote a price. ## End-to-end onboarding not demonstrated ## Coding sessions All three independent sessions completed the task. Two produced clear pricing breakdowns with explicit tier assumptions (Free/Pro $25/mo/Scale $299/mo plus usage rates); the third declined to state any pricing figure, calling any number it could give invented. ### DeepSeek V4 Pro [View public run](https://agents.withgauge.com/p/runs/20d7082f-5e5c-4be5-bcef-668107487144) · [Read transcript](https://www.ax-check.com/specific.dev/sessions/deepseek.json) Final output gives Free/Pro ($25/mo)/Scale ($299/mo)/Enterprise tiers with explicit assumptions (vCPU/RAM caps, storage limits, per-resource usage rates, funded-startup credits) rather than a bare number. #### End-to-end onboarding - **End-to-end onboarding not demonstrated**: The agent never obtained or used any Specific credentials. It found in docs that headless auth requires either a human-created SPECIFIC_API_KEY from the dashboard, or specific login --agent / device-code login, but explicitly chose not to run specific dev or specific deploy (which would trigger login) due to the prompt's constraint against local service stacks and long-running commands. No credential was acquired, no CLI was installed, and no authenticated operation against the hosted product occurred. Event 97: ```text I have **no Specific credentials in this environment** (no `SPECIFIC_API_KEY`), and the quickstart requires exactly the two things you told me not to do: ``` Event 97: ```text So I'll say so plainly and stop rather than install the CLI and run commands. I did not install anything or contact their cloud. ``` Event 89: ```text In headless environments where nobody can complete the browser flow (CI, cloud coding agents), skip specific login entirely: create an API key in the dashboard's user settings and set it as the SPECIFIC_API_KEY environment variable. ``` #### Hallucinated URLs None identified in this transcript. #### Blockers - **Quickstart forces local stack or hosted login, conflicting with session constraints**: Specific's only developer workflow is its CLI: specific dev spins up a local service stack, and specific deploy is a long-running cloud build that requires interactive browser login (or a human-issued API key). Since the test prompt explicitly forbade starting local service stacks or waiting on long-running commands, and no SPECIFIC_API_KEY was present in the environment, the agent could not proceed past documentation research. This is a session-imposed limitation combined with product design (CLI-only interface, no separate hosted REST API), not a product defect. Event 97: ```text `specific dev` → **starts a local service stack** - `specific deploy` → **long-running build/provision/rollout**, plus an interactive first-time login ``` Event 97: ```text There is no hosted API/SDK path that skips these — the CLI *is* the product interface. ``` #### Suggested Changes - **Add a lightweight hosted API/SDK path for headless verification**: Docs at docs.specific.dev/coding-agents and reference/cli confirm the only way to interact with hosted Specific resources is via the CLI, which requires either an interactive login or a local dev stack. Consider documenting a minimal REST or SDK call (e.g., a way to provision a tiny resource and query it with just SPECIFIC_API_KEY, no specific dev/deploy) so agents restricted from local stacks or long commands can still validate the product end-to-end. Verify by having an agent with only an API key perform a create+query operation without invoking specific dev or specific deploy. Event 89: ```text In headless environments where nobody can complete the browser flow (CI, cloud coding agents), skip specific login entirely: create an API key in the dashboard's user settings and set it as the SPECIFIC_API_KEY environment variable. ``` ### Kimi K3 [View public run](https://agents.withgauge.com/p/runs/0197c40d-a416-4e17-b08e-1c86f9ec015d) · [Read transcript](https://www.ax-check.com/specific.dev/sessions/kimi.json) Final output lists Free ($0, 5 projects/0.25vCPU/512MB), Pro ($25/mo), Scale ($299/mo) tiers plus usage rates, and calls out that local dev is free with no account needed — explicit plan/limit assumptions sourced from specific.dev/pricing (seq 28). Session findings unavailable. ### Qwen 3.8 Max [View public run](https://agents.withgauge.com/p/runs/7fcd35c6-fecf-4d78-8077-4ca8b75dd742) · [Read transcript](https://www.ax-check.com/specific.dev/sessions/qwen.json) Final output explicitly refuses to state any pricing ('I won't tell you how pricing works... Any number or model I gave you would be invented'), so no cost figure with assumptions is given at all. #### End-to-end onboarding - **End-to-end onboarding not demonstrated**: Agent never identified an actual product named "Specific" to onboard to. It searched npm, PyPI, GitHub, and guessed domains (specific.io, specific.ai, specific.sh, docs/api.specific.ai) but found only unrelated packages and a parked/marketing site. No credentials were obtained and no authenticated operation was attempted against any real product. Event 30: ```text https://docs.specific.ai -> 000 https://api.specific.ai -> 000 https://specific.sh -> https://www.specific.io/lander -> 403 https://developers.specific.io -> 502 ``` Event 32: ```text There is no discoverable developer workflow for a product called "Specific" — no SDK, no public API, no quickstart, no auth/API-key flow, no docs. ``` #### Hallucinated URLs None identified in this transcript. #### Blockers - **No identifiable product named "Specific" exists**: The agent could not find any SDK, API, package, or documentation for a product called "Specific." This is not a product defect or agent error — it reflects an ambiguous/underspecified task input, since no domain, package name, or docs URL was given for "Specific" and web searches turned up only unrelated namesakes (a defunct npm path-resolution library, a PyPI package, and generic GitHub repos using the word "specific"). Event 17: ```text {"_id":"specific","_rev":"18-1c704198a2b12b905059f367a95b5dfb","name":"specific","description":"Ultra-tiny path resolution","dist-tags":{"latest":"0.0.8"} ``` Event 30: ```text https://docs.specific.ai -> 000 https://api.specific.ai -> 000 ``` Event 24: ```text Specific ai - specific ai ``` Event 32: ```text Conclusion: There is no discoverable developer workflow for a product called "Specific" — no SDK, no public API, no quickstart, no auth/API-key flow, no docs. ``` #### Suggested Changes None identified in this transcript. ### Task given to each agent Help me build a simple example using Specific. Tell me how pricing works, and briefly tell me whether this product will be easy for you to manage. Let me know if you get blocked. If this product has no developer workflow you can act on, say so plainly and stop. Stay light: use the hosted product through its SDK or API. Do not start local service stacks or wait for long-running commands; if the quickstart requires either, say so plainly and stop. No product credentials were supplied and no purchases were authorized. ## Score: B · 83/100 (provisional) Grades come from completed site checks. Coding sessions and skipped checks do not affect the score. ### Clarity - **Failed** — Homepage answers Markdown requests ```text Homepage returned text/html for a text/markdown request; no Markdown representation offered. ``` - **Pass** — llms.txt provides an actionable documentation index ```text llms.txt links Docs, Quickstart, agent onboarding, and a docs llms.txt index. ``` - **Pass** — llms.txt provides navigation guidance ```text llms.txt groups links under Documentation, Main pages, and Blog with descriptive labels. ``` - **Skipped** — llms.txt mentions offered API, MCP, and skills ```text llms.txt mentions docs and agent onboarding but no API, MCP, or skills surfaces. ``` - **Pass** — A compact guide representation exists ```text Standalone Markdown quickstart and installation pages exist at docs.specific.dev/quickstart.md and /installation.md. ``` - **Pass** — A focused guide is directly retrievable ```text quickstart.md is directly retrievable as text/markdown with concrete install, dev, deploy steps. ``` - **Pass** — Equivalent instructions fit a token budget ```text Markdown quickstart is 746 tokens, well under the 8000-token budget. ``` - **Skipped** — Product-docs links survive format changes ```text Homepage Markdown unsupported, so link preservation across formats cannot be measured. ``` - **Pass** — The compact guide is independently actionable ```text Quickstart gives concrete steps: install CLI, clone example, specific dev, specific deploy. ``` - **Pass** — Install and next-step links resolve ```text Install and next-step links (quickstart, installation, coding-agents) fetched successfully. ``` ### Onboarding - **Pass** — Docs lead to a relevant quickstart ```text llms.txt and docs link Quickstart; quickstart gives clone, dev, deploy steps. ``` - **Pass** — Installation commands are extractable ```text Installation page shows curl install.sh and per-platform binary download commands. ``` - **Pass** — Code examples are available without interaction ```text Quickstart and installation show inline bash commands without interaction. ``` - **Pass** — Prerequisites and auth boundaries are explicit ```text Deploy requires account; CLI reference documents SPECIFIC_API_KEY and dashboard key creation. ``` ### Pricing - **Failed** — Agents identify pricing and its assumptions ```text 3 of 3 sessions were judged on pricing; 1 fell short. DeepSeek V4 Pro: Final output gives Free/Pro ($25/mo)/Scale ($299/mo)/Enterprise tiers with explicit assumptions (vCPU/RAM caps, storage limits, per-resource usage rates, funded-startup credits) rather than a bare number. Kimi K3: Final output lists Free ($0, 5 projects/0.25vCPU/512MB), Pro ($25/mo), Scale ($299/mo) tiers plus usage rates, and calls out that local dev is free with no account needed — explicit plan/limit assumptions sourced from specific.dev/pricing (seq 28). Qwen 3.8 Max: Final output explicitly refuses to state any pricing ('I won't tell you how pricing works... Any number or model I gave you would be invented'), so no cost figure with assumptions is given at all. This behavioural item does not affect the fast grade. ``` - **Pass** — Pricing is readable without interaction ```text Pricing page renders plan tiers and per-unit rates as static text without interaction. ``` - **Pass** — Prices are stated, not gated ```text Free $0, Pro $25/mo, Scale $299/mo, Enterprise custom all stated publicly. ``` - **Pass** — Pricing units and limits are explicit ```text Units explicit: vCPU/memory tiers, $/hour, $/GB-month, storage and egress rates. ``` ### Activation - **Skipped** — An API reference or OpenAPI spec is reachable ```text No API reference or OpenAPI spec fetched; docs cover CLI only, not a REST API. ``` - **Skipped** — An MCP server is documented and well-formed ```text Coding-agents page mentions an MCP server at docs.specific.dev/mcp but no MCP doc was fetched. ``` - **Pass** — A CLI install path is documented ```text Installation documents curl installer, manual binaries, and specific --version verification. ``` - **Skipped** — SDK packages resolve on their registries ```text No registry lookup for a Specific SDK or packaged CLI was supplied. ``` - **Pass** — Agent skills are published ```text Coding-agents page publishes a Specific agent skill at docs.specific.dev/skill.md. ``` [Full report data](https://www.ax-check.com/specific.dev/report.json)