Mistral Vibe CLI vs Verdent

Mistral Vibe CLI 84

Framework reading, adapted to a harness driven by a pipeline. --enabled-tools and --disabled-tools with globs and regex (in -p the first disables everything else), per-tool and per-MCP-tool permissions, and switches for subagents and background processes (20 of 25). --max-turns, --max-tokens, --max-price and --time-limit, with text, json and streaming output and an export file with usage and cost (20). Exit codes 0 to 4 with named outcomes, refusals reported with a code, and config problems listed under warnings (18). --resume SESSION_ID and --continue, SIGTERM stops a run so it can be resumed, and model API failures are retried before exit 2 (16). One required input (the prompt), a GitHub Action and the ACP server, but no SDK, and the app-server's JSON-RPC protocol isn't documented for outside use (10). 20+20+18+16+10 = 84.

Verdent 35

Harness reading. MCP servers per project or globally, subagents, skills and rules, but no per-tool disabling or deferred loading found (10). No output-size, turn-limit or budget controls for a run were found beyond credit balances (5). No exit codes or error format, since there is no headless mode (3). Parallel workers in git worktrees and a review column make runs reviewable and reversible, with no resume by ID documented (10). No CLI, API or SDK. A pipeline can only reach Manager through a paired Telegram, Slack or Feishu bot (7). 10+5+3+10+7 = 35.

Mistral Vibe CLI 60

Harness reading of the published rubric, as for openai-codex and qwen-code. No payment protocol (0). Plan prices (Pro $14.99 a month, Team $24.99 a user a month) and per-token API prices for the default model ($1.5 and $7.5 per million) are public without a login (20). A Free plan includes limited coding sessions, and the docs document a fully offline set-up with a local model and no Mistral account. Whether the Free plan asks for a card wasn't checked (20). An agent can install the CLI and run it against a local OpenAI-compatible server with no sign-up, since a llama.cpp provider on 127.0.0.1:8080 is in the defaults (20). 0+20+20+20 = 60.

Verdent 20

Harness reading of the published rubric, scored on the Verdent subscription the agent needs. No payment protocol (0). Plan prices from $5 a month, top-up prices and the credit value of about $0.059 are public without a login (20). Free Mode and the free trial ended on 29 October 2026, so there is no free tier (0). A person signs up in a browser (0). 0+20+0+0 = 20.

Mistral Vibe CLI 85

Framework reading. A configuration reference with each key's type and default, the export.json fields, and the --agent-socket JSON-RPC methods with typed params and results in the README, plus ACP. Since 3.0.3 vibe config schema prints a JSON schema of every setting, including each built-in tool's; the streaming events still have no published schema (20 of 25). docs.mistral.ai/llms.txt indexes the CLI pages, each with a Markdown twin (10). The safety page says what each agent allows and when to use it, and on 2 November it agrees with the source. -p runs the configured default_agent and denies calls that need approval, and the offline guide and configuration reference name enable_update_checks. The 18 October contradictions and TODO notes are gone (15 of 20). Output formats, agent names, permission values and OpenTelemetry redaction modes are enums, and the config is validated on load (12). Many examples and an exit-code table that maps each code to the outcome names in export.json (13). Dated changelog and version tags (15). 20+10+15+12+13+15 = 85.

Verdent 55

Framework reading for a harness. No schema for settings or output, only mcp.json examples in the standard mcpServers shape (5). Every current docs page has a Markdown twin at the same path with .md, though no llms.txt (10). The docs say when to use Agent Mode or Plan Mode and what Plan Mode can't do, but the Agent Mode page lists the actions that need approval only as "certain tool executions" and "destructive operations" (12). Modes and BYOA agent types are enumerated, permissions are not (5). Examples on the MCP, BYOA and messaging pages, and a recovery page, with no error codes (8). Version numbers and a dated public changelog (15). The archived 1.0 docs are still served beside the current ones. 5+10+12+5+8+15 = 55.

Mistral Vibe CLI 69

Local-package reading. PyPI, uv, pip, Homebrew at 3.0.4 and an install script, with Python 3.12 or later stated and UNIX the supported target (20). A public CI workflow on push and pull request to main runs pre-commit, the test suite and a build and install smoke matrix, but GitHub's pages and API were refused in this session, so whether main is passing is unchecked (15 of 25). The 18 October issue counts came from shields.io badges, a third-party copy of GitHub data this run bars, so they are struck. Every release has a Fixed section in the changelog, and how many open issues are crashes or regressions, and how fast they're answered, couldn't be read (10 of 25). The changelog states Keep a Changelog and semver and marks breaking changes, but the hooks rename shipped in minor 2.21.0, patch 2.11.1 changed the -p approval default, patch 2.26.1 removed the bundled local model, and patches 3.0.3 and 3.0.4 carry marked config, model and session breaks (9). 3.0.4, classified Production/Stable (15). 20+15+10+9+15 = 69.

Verdent 43

Local-package reading, as for the other closed-source harnesses. The desktop app installs from verdent.ai with macOS 11 or later and 64-bit Windows 10 and 11 stated, and the extension from the editor marketplaces, but not from a package registry (15). No public CI or test suite, since the source isn't published (0). No public issue tracker, every task runs through Verdent's servers, and no status page was found, so there is no readable record (5). A dated changelog by version, with the end of Free Mode called out in v2.18.0, but no breaking-change sections (8). 2.18.0, generally available (15). 15+0+5+8+15 = 43.

Mistral Vibe CLI 81

Apache-2.0, with LICENSE and pyproject in agreement (30). The README sends readers to Mistral's privacy policy (effective 3 September 2026), which names Vibe. The commercial terms (effective 25 September 2026) say no training on customer data unless a product's default or an opt-in allows it, and the pricing page lists model training as opt-out on Free, Pro and Team. The configuration reference calls telemetry anonymous without listing its fields. The offline guide now names enable_update_checks = false with enable_telemetry = false, as the source reads them (16 of 30; 14 on 18 October, when it named a key the source defines but never reads). Removals are listed in the changelog, deprecated settings are named in the configuration reference and warned about at startup, the agent socket may change without notice, and custom Python tools and the bundled local model were removed in patch releases with warnings (10 of 20). Telemetry is disclosed with an opt-out in the README and the configuration reference, and the README names the server-managed shell-tool rollout, but its fields are only in the source (14 of 20). 30+16+10+14 = 70.

Verdent 55

Proprietary, under terms from Verdent AI, Inc. that are clear on use and ownership (15). The privacy policy and security page agree that data stays in the US and that input isn't trained on without consent, and the security page lists subprocessors. The privacy policy gives no retention periods, names MiniMax as a provider the security list leaves out, and deletion is by email request (14). The credit bonus carries a promise of 15 days' notice, but Free Mode was retired on the day v2.18.0 shipped and no deprecation policy was found (6). Subprocessors and their locations are listed, and the domains the app calls are named, with event logging undescribed (12). 15+14+6+12 = 47.

Mistral Vibe CLI 58

Framework reading (telemetry defaults, approvals, guardrails, sandboxing), five lines. Telemetry and crash reporting are on by default with an opt-out, and the source sends tool names, approval decisions, the model and file counts to api.mistral.ai only when a Mistral key is set. Telemetry also turns on GrowthBook experiments, and one of them (vibe_cli_smart_approve_default) can make smart-approve the default. Credentials are Mistral keys in an environment variable, a plaintext ~/.vibe/.env or the keyring, workspace-scoped with optional expiry per the mistral-api listing, and MCP servers can use OAuth (12 of 30). Five agents with a read-only plan agent and bash allow and deny lists, a prompt for access outside the working directory, outside_workdir = "deny" (3.0.3) to refuse it outright, an admin enable_auto_approve = false (3.0.1) that forbids auto-approve, and pre_tool hooks that can deny calls. The default auto-approves file edits, and there's no sandbox on the host (13 of 20; 11 on 18 October, before the two settings). Project config, hooks and MCP servers load only from trusted folders, and -p needs --trust to load them. The safety page warns that smart-approve isn't a sandbox. No network egress control on the host, and MAI-2026-003 says injected instructions could trigger the bypassed commands (8 of 15). OpenTelemetry tracing is opt-in, with default, strict and no redaction, and each run can write a session journal and an export with the effective config. Managed admin config exists, and audit logs come with the Enterprise plan (13). SECURITY.md asks for private reports, security.txt is valid (expires 5 May 2027) and links a HackerOne form, and Mistral publishes dated advisories with CVEs, three to 2 November. No bounty terms or certification report were read, so 12, as mistral-api and mistral-voxtral-transcribe take on the same Mistral pages (14 on 18 October). 12+13+8+13+12 = 58.

Verdent 33

Framework reading (telemetry defaults, approvals, guardrails, sandboxing). Conversation history and tool results go to Verdent's US servers and on to Azure, Bedrock or Vertex, with zero data retention stated for Vertex and Bedrock only. Event logging to log.verdent.ai has no documented opt-out. BYOA keys stay on the machine (10). Plan Mode is read-only, Agent Mode asks before destructive commands and edits outside scope, and workers are isolated by git worktree, but there is no sandbox or network control and the approval list isn't enumerated (9). No prompt-injection guidance found, while web search and fetch results enter the context (2). Usage history in the dashboard, no audit log found (5). A security page with a disclosure address and a five-day acknowledgement, SOC 2 and ISO/IEC 42001 being pursued rather than held, no security.txt and no bug bounty (7). 10+9+2+5+7 = 33.