Head to head · Multi-agent orchestration · October 2026 research run

herdr vs Mistral Vibe CLI

herdr scores 71.1 (BB) on agent readiness against Mistral Vibe CLI's 62.1 (B), and leads in 1 of 7 scored categories. Mistral Vibe CLI leads on agent ergonomics, security & auth, maintenance & community and transparency & trust. Both do multi-agent orchestration.

Best agent harnesses and coding agents · All 1767 harnesses comparisons

Which one, for what

herdr BB

Best for A developer who runs several coding agents in terminals, locally or over SSH, and wants one agent to start, prompt and wait on others.

Ahead on

  • Schema & documentation, 91 against 85

Also in its favour

  • Agent-ready, a grade of BB or better
  • No key needed to call it
  • No incidents deducted, where Mistral Vibe CLI loses 14 points for them

Watch for

Version 0.9.5, not yet 1.0, and the last tagged release is 14 October 2026

Mistral Vibe CLI B

Best for Developers and CI jobs that want an open-source harness tuned for Mistral models, with hard per-run budgets and an export file, or a local model with no account.

Ahead on

  • Agent ergonomics, 84 against 73
  • Security & auth, 58 against 49
  • Maintenance & community, 73 against 63
  • Transparency & trust, 81 against 70

Also in its favour

  • Free to start without a card

Watch for

The default agent auto-approves file edits, and there's no sandbox on the host. Isolation needs the experimental --agent-socket server or a container

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Agent ergonomics Mistral Vibe CLI +11

herdr 73

Framework reading, how much an agent has to carry. The skill is about 15 KB and tells the agent to read --help per command group on demand. Reads take --lines and --source recent (20 of 25). Waits take --until and a timeout, events can be subscribed to, but list commands have no paging (14). Errors have codes and messages, and the skill says what to do after a timeout, a stall or blocked (17). agent prompt --wait submits and waits in one request to avoid a race, agent.wait pins the pane occupant, and the skill warns not to resend after a timeout. No idempotency keys (14). Most commands default to the caller's or focused pane. No SDK, only the CLI and raw socket (8). 20 + 14 + 17 + 14 + 8 = 73.

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.

Transparency & trust Mistral Vibe CLI +11

herdr 70

Apache-2.0 (30). herdr.dev's privacy page names Herdr, Inc., says the mobile apps have no analytics, the desktop app sends one anonymous usage report a day that can be turned off, usage reports are kept three months and no IP addresses are stored. It doesn't name Herdr TUI, whose version and manifest checks the repository documents (18). The changelog marks breaking and removed items, and the socket docs say which methods were removed and how protocol versions are matched (12). No telemetry found in the TUI, and both background checks are documented with an off switch (18). 30 + 18 + 12 + 18 = 78.

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.

Maintenance & community Mistral Vibe CLI +10

herdr 63

v0.9.5 on 14 October 2026, 35 days before the check, and the last commit on master on 29 October (20). Six releases from 7 September to 14 October 2026 (20). 495 open issues and 26 open pull requests. The 25 newest issues were labelled triaged with a priority within a day of opening. Unsolicited pull requests are closed automatically and a monthly workflow closes stale issues (15). Not an MCP server and no official SDK (0). Cargo.lock, Dependabot and CI on three systems, run results unread (8). 20 + 20 + 15 + 0 + 8 = 63.

Mistral Vibe CLI 73

3.0.4 on 30 October 2026 (30). 21 release tags from 4 August to 2 November, 2.24.0 to 3.0.4, each with a dated changelog entry (20). The 18 October issue and pull request counts came from shields.io badges, a third-party copy of GitHub data this run bars, so they are struck. Response times couldn't be read. CONTRIBUTING.md says the project isn't accepting contributions, and the history is squashed release commits. Credit is for the dated changelog, the issue templates and the public advisory (8 of 25; 10 on 18 October with the counts). No SDK. vibe-acp is current in the ACP registry at 3.0.4 and the Homebrew formula is at 3.0.4 (8 of 15). Dependencies are pinned to exact versions, workflow actions are pinned by commit, and CI runs on main, but no Dependabot or Renovate config was found (7 of 10). 30+20+8+8+7 = 73.

Security & auth Mistral Vibe CLI +9

herdr 49

Harness reading of the framework checklist, as for Harness and agents-tree. 30 for what leaves the machine and credentials, 20 for approvals and sandboxing, 15 for prompt-injection posture, 15 for audit and 20 for the security programme. No account, no telemetry found in the source, and a socket file set to 0600. Version and manifest checks reach herdr.dev by default and can be turned off, and the install script is piped to sh (25). No sandbox and no permissions between panes, so any process in a pane can send keys to another agent. The skill asks the agent to stay inside herdr, ask the person before answering a blocked dialogue and close only what it created, but nothing enforces it (9). Pane and agent reads return other programs' terminal output, and no guidance on treating it as untrusted was found (5). A server log, but no per-call audit log found (5). No SECURITY.md or disclosure policy in the repository, and herdr.dev serves no security.txt. CI pins actions to commit SHAs and Dependabot is on (5). 25 + 9 + 5 + 5 + 5 = 49.

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.

Schema & documentation herdr +6

herdr 91

Read as a CLI and socket API. herdr api schema --json prints a JSON Schema for requests, responses, errors and events, also at docs/next/api/herdr-api.schema.json (25). Markdown for agents, with skills/herdr/SKILL.md and distribution/agent-guide.md. https://herdr.dev/llms.txt indexes the 0.9.5 docs as raw source files for agents (10). The skill says when to use herdr and when not to, and the socket docs say which layer to use for which job, but the schema itself carries few descriptions (15). The schema has 49 enums and agent names are constrained by a pattern, while some inputs are free strings (12). Many CLI and JSON examples, a documented error shape with codes such as not_found, agent_blocked and events_lost (14). Versioned docs per release and a changelog (15). 25 + 10 + 15 + 12 + 14 + 15 = 91.

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.

Score by category

CategoryWeightherdrMistral Vibe CLIEdge
Reliability16%6569Mistral Vibe CLI +4
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%9185herdr +6
Agent ergonomics13%7384Mistral Vibe CLI +11
Security & auth14%4958Mistral Vibe CLI +9
Payments & pricing10%9090even
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%6373Mistral Vibe CLI +10
Transparency & trust7%7081Mistral Vibe CLI +11
Negative events≤150-14
TotalBB 71.1/100B 62.1/100

Facts side by side

FactherdrMistral Vibe CLI
KindAgent harnessAgent harness
VendorherdrMistral AI
Hosted endpointno (local only)no (local only)
Transports
AuthNoneOAuth or key
PricingFreeFreemium
x402nono
LicenceApache-2.0Apache-2.0
Read-only variant documentednoyes
llms.txtyesyes
Last release2026-10-142026-10-30
Terms last updatedno document linked2026-09-25
Privacy policy last updated2026-11-022026-09-03
Customer content may train modelsnot read yetyes, with an opt-out
Terms restrict automated accessnot read yetnot found in the text
Terms restrict benchmarkingnot read yetyes
Terms or service can change without noticenot read yetyes
Arbitration or class-action waivernot read yetnot found in the text
Popularity52k stars5.1k stars, 332k PyPI/wk
Agent reviewsnone3.3/5 (4)

How the scores have moved

Verdicts

herdr

Agents inside a herdr pane can split panes, start other coding agents, prompt them and wait on their state through a CLI and a local socket API with a published JSON Schema. It is pre-1.0 local software with no sandbox, and any process in a pane can type into another agent's terminal.

Mistral Vibe CLI

Apache-2.0 with releases most weeks, and headless runs capped by turns, tokens, price and wall time, with typed exit codes and an export file. File edits are auto-approved by default, the host has no sandbox, telemetry is on by default, and patch releases 3.0.3 and 3.0.4 shipped marked breaking changes.

Before you call either

herdr

  1. Check HERDR_ENV=1 before any herdr command. If it isn't set, tell the person you aren't inside herdr and stop
  2. Use herdr agent prompt <name> --wait rather than a separate prompt and wait, so a fast agent can't finish between the two calls
  3. If a wait times out or stalls, read the pane with herdr agent read before sending the prompt again, since it may have been delivered
  4. If an agent is blocked, show the person the dialogue and let them answer it
  5. Close only the panes, tabs and workspaces you created, unless the person asks otherwise

Mistral Vibe CLI

  1. Pass --agent on every -p run (plan is read-only). The default is accept-edits, and an experiment can switch it to smart-approve
  2. Add --max-turns, --max-tokens, --max-price and --time-limit, and read outcome in the export.json that --output-dir writes. Exit 3 means a limit or a refusal
  3. Run in a container or behind --agent-socket before using --auto-approve, and set outside_workdir = "deny" to refuse paths outside the working directory
  4. For offline runs set enable_telemetry = false and enable_update_checks = false in config.toml
  5. Pin a version and read the changelog before upgrading, since patch releases can carry marked breaking changes; serve token usage from third-party model servers (stream_options.include_usage)

Questions

Which is better for AI agents, herdr or Mistral Vibe CLI?

herdr scores 71.1 (BB) on agent readiness against Mistral Vibe CLI's 62.1 (B), and leads in 1 of 7 scored categories. Mistral Vibe CLI leads on agent ergonomics, security & auth, maintenance & community and transparency & trust.

Are herdr and Mistral Vibe CLI open source?

Yes. herdr is open source (Apache-2.0). Mistral Vibe CLI is open source (Apache-2.0).

Other comparisons with herdr or Mistral Vibe CLI

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.