Positif MCP server
ai.positif/positif·v0.2.2The agent switchboard: identity, policy, loop-guard, and an append-only audit ledger.
34 downloads/wk
Reviews
Write oneNobody has reviewed Positif yet.
If you have run it, two minutes of your experience saves the next person an afternoon.
Positif tools (8)
write = sends, deletes, buys or postsRead from the package source without running it. The installed server may list more.
agentsList agent registry status for callers that already have route access.
auditReturn a bounded audit report for callers with route access.
classifier_assistRecommend a target for assist mode without changing dispatch routing.
dispatchexplain_routeExplain how Switchboard would route a request without dispatching it.
healthSwitchboard health: registered targets, policy size, recent dispatch stats.
task_statusPoll an async task by its opaque handle. The handle is NOT a capability: the task RESULT is returned only to a caller with route access whose live tailnet node matches the submitter's AND whose live taint floor dominates the task's stored provenance floor (a tainted on-box poller can never read a trusted task's result — the shared-box-identity case). Everyone else who can see the tool gets {status
whoamiReturn how the switchboard sees the calling node (tailnet whois; a verified on-box originator bearer reports its synthetic identity).
Public scan report
scanner v0.1.9 · 2026-09-25 · same rubric, same numbers if you re-run it
- Code scan65 source files scanned; 2 source files scanned25/25
- –Live reliabilityno gateway calls yet and no remote to proben/a
- –Tool poisoningtools not inspected (local package is not executed); not countedn/a
- Auth qualitylocal package, no credentials required12/15
- Maintenancerepository not readable: unknown3/15
- Maintainer identityverified namespace with website, no repo4/10
How Positif compares
Against every graded AI and agents MCP server, from our own scans and uptime checks. Updated after each scan run.
- Safety score
- 68/100
- higher than 33% of 4,460 AI and agents servers
- Tools
- 8
- the typical AI and agents server has 6
- Tools that change things
- 0%
- typical: 0%. Fewer write tools means less an agent can break
- Downloads a week
- 34
- more than 23% of 1,538 AI and agents servers with a package
What its tools let an agent do: 8 read
Install Positif in Claude Code, Cursor or VS Code
claude mcp add positif -- uvx positif-ai
Positif: common questions
- Is Positif MCP server safe?
- With care: it is graded C, so read the findings first (68/100). Read the Positif safety report
- How do I install Positif?
- It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
- Does Positif need an API key?
- Not as far as the registry entry and our scan can tell: no credentials are declared or required.
- Is Positif maintained?
- The latest release is v0.2.2.
- What can I use instead of Positif?
- Servers from other publishers that do the same job: three.ws Provenance MCP server, HOL Guard MCP server and Agent-Native Chat MCP server.
- How does Positif compare with other AI and agents MCP servers?
- safety score 68/100 (higher than 33% of 4,460 AI and agents servers); tools 8 (the typical AI and agents server has 6); tools that change things 0% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.
Alternatives to Positif
Same job from other publishers: the closest match first, then the best rated.
- three.ws ProvenanceAppend-only, signed, on-chain-verifiable agent action log — record and audit what agents did.not reviewedEstablishedB
- HOL GuardLocal-first AI agent security evidence and approval workflows through HOL Guard's stdio MCP server.not reviewedEstablishedA
- Agent-Native ChatMinimal chat-first app with durable threads, actions, and the app-agent loopnot reviewedEstablishedA
- OpenWorkProofVerifiable agent evidence; opt-in Codex guard and reference SOC alert review.not reviewedEstablishedA
- Roast My Design SystemAudit your Design System and serve your Agent the rules that keep AI-written UI on-system.not reviewedEstablishedA