AI & Agents

Planwright is MCP-native. Agents don't receive webhooks — they call tools.

When an agent session starts, it calls planwright_set_repo to select a project, then planwright_list_objectives to see what's scheduled. It claims an objective with planwright_claim_objective, posts a decomposition plan with planwright_append_plan, makes code changes, and calls planwright_request_acceptance when done. Every state change flows through the MCP server, which validates the transition, appends a signed audit record, and notifies the human reviewer. Agents never mutate board state directly.

Agents operate within a strict permission boundary. Some things are intentionally off-limits.

Agents can propose objectives and initiatives — humans decide what gets scheduled.

An agent can call planwright_create_objective to propose new work based on what it observes in the codebase: a missing test, a performance regression, a gap in the API surface. Proposed objectives land in the Backlog lane and stay there until a human schedules them. Agents can also call planwright_create_initiative to suggest a new grouping for related objectives. Initiative generation is most useful mid-execution: an agent working on one objective may discover a natural cluster of follow-on work and propose it as a named initiative so the human can review and schedule it as a set.

Every objective must carry a strategic classification before it can be claimed.

When an agent calls planwright_claim_objective, it must provide a strategic_focus field if the objective has no existing classification. The taxonomy is four buckets — BUILD, GROW, TRUST, SCALE — each with specific areas. Agents are expected to provide a confidence score from 0 to 1. Low confidence is fine: the system records it and surfaces it for human review. An objective with no classification cannot be claimed. This ensures every piece of work can be traced to a strategic intent, and that the portfolio allocation data in the Strategic Focus report is always complete.

Agents follow a branch strategy returned at claim time and attach trailers to every commit and PR.

The planwright_claim_objective response includes a branchStrategy field — either branch-pr or direct. The agent must follow it. Every commit must include both the Planwright-Objective and Agent-Run trailers so Planwright can attribute changes to the correct objective and run. Every PR opened by an agent must include the Planwright-Objective trailer in its body so the diff is linked automatically via GitHub webhook.