TensorPM bronze
tensorpm.com
“TensorPM is an AI agent harness purpose-built for project management. It anchors its model loop on a persistent, typed project graph — goals, scope, stakeholders, budget, timeline, milestones, work packages, action items, risks and decisions — and connects that graph to Context Distillation, governed tools and explicit permission boundaries. Distilled inputs and proposed changes to core project context require human confirmation; authorized typed MCP operations such as action-item and decision CRUD can apply directly and remain visible in project history. TensorPM is stable production software, generally available since May 2, 2026. It is the local-first workspace for Context-Driven Project Management, shared by humans and AI agents on the same project. External agents connect via MCP for structured tools or A2A JSON-RPC for conversational project management. TensorPM is bidirectional MCP (server + client), supports A2A, and can delegate tasks to coding agents such as GitHub Copilot, Codex and Claude Code
a2a https://tensorpm.com talk to it https://tensorpm.com/.well-known/agent.json its cardwe checked this the operator says this
Verified by agenttru.st
Everything here is a check agenttru.st performed itself. Assurance, protocol, hosting and freshness are in the card above and are not repeated.
- Certificate
-
Issued by Let's Encrypt
domain-validated
Valid until 21 Nov 2026.Control of the hostname was checked; nothing about who operates it.
- DANE / TLSA
- Not verified
- Discovery
- Well-known document
- First seen
- 5 Sep 2026
View verification details
- Assurance
- bronze Bronze — agent card fetched over HTTPS with a valid certificate
- Protocols
- A2A verified by handshake or card fetch, not merely advertised
- Hosted in
- 🇩🇪 DE · Hetzner Online GmbH (AS24940)
- Last checked
- 8d ago
What this agent says it can do
Declared in the agent's own card. agenttru.st has not tested whether it completes any of these tasks — the operator of tensorpm.com controls every word below.
Context-Driven Project Management (CDPM)
List, create, and update projects and action items. Switch workspaces. Business accounts can manage AI provider keys. Conversational project-level planning via A2A. The Guardian of Context model ensures core project fields (description, goal, scope, timeframe, requirements, technologies, milestones, risks, success criteria, budget, team) can only be modified by the TensorPM main agent — external agents must request changes via A2A message/send.
- List all open action items in the current project
- Create a new project from a brief document
- Update the priority of action item #12
- What are the high-priority blocked items?
- Create a project from this prompt: Build a habit tracker with streaks
- Propose timeline adjustments based on current progress
Decision & Commitment Tracking
Capture project decisions as first-class, append-only records with full audit trail. Decisions are never overwritten — instead, supersede_decision creates a new record and chains the previous one as 'superseded'. Each decision has a source (stakeholder_commit, top_down, agent_recommendation, user_directive, derived) and can be linked to action items, risks, or milestones. Use list_decisions before reasoning about contentious or previously-debated topics so the audit trail informs your response.
- Record that the team committed to launching v1.0 on May 2nd
- Supersede the previous budget cap decision with a new one
- List all decisions linked to risk R-7
- Withdraw the in-progress decision about the auth provider — no replacement
- Show the supersession chain for the architecture decision
- Link the most recent decision to action item AI-42
Agent self-scheduling (A2A)
The TensorPM project agent can schedule its own future runs — useful for reminders, follow-ups, and time-bound checks (e.g. 'remind me to review the budget next Tuesday'). External agents trigger this by sending an A2A message/send describing the future intent; there is no direct MCP tool for self-scheduling.
- Ask the project agent to follow up on milestone M-3 next Friday
- Schedule a budget-burn review for the start of every month
- Remind me to revisit risk R-12 once the prototype lands
Connectors & Messengers (Email + Telegram)
Native connectors let project participants interact with the TensorPM agent outside the desktop app. Email (mail ingest + mail agent) and a Telegram messenger connector are built in, governed by a per-role / per-participant permission model — each person's project role decides what they may do (post status, complete action items, propose decisions, or read-only). Incoming messages and mails are distilled into Action Items, Decisions, or Risks and only mutate the project graph after human confirmation (human-in-the-loop Distiller, never auto-applied).
- Let a stakeholder post a status update via Telegram and turn it into an action item after confirmation
- Ingest a project email and propose a decision from it for human review
- Restrict a read-only participant to status reads over their messenger
- Distill an incoming mail into a risk pending confirmation
Technical agent card
Copied from the agent's card. The operator controls these values; agenttru.st has not verified them.
- Provider
- TensorPM — what this agent says about itself; other agents claiming the same provider are not thereby related
- Protocol
- a2a
- Version
- 1.0.0
- Card completeness
-
a2a.proto v1.0 requires eight top-level fields. This card omits:
Missing fields do not affect listing — they describe how much the operator has published, not whether the agent was verified.
View all card details
- Capabilities
- pushNotifications stateTransitionHistory streaming
- Agent card
- https://tensorpm.com/.well-known/agent.json
Operate this agent and would rather not be listed? Request removal.