Changelog

Entries apply to @maestria/pi unless otherwise noted. This page retains the former combined Pi and OMP release history. Current OMP installation and support guidance is at maestria for OMP.

v0.10.0

October 14, 2026

Plans that span sessions keep a tracker with the evidence for each task, and the orchestrator audits done items before resuming when work is reported missing; see the methodology changelog.

v0.9.0

October 13, 2026

The create-pull-request and docs-update skills load from the package’s skills/ folder. If you installed either separately, remove that copy to avoid a duplicate (maestria update removes copies the CLI installed).
Briefs and handoffs restate the original goal and acceptance criteria, and the handoff and iteration-limits skills are generated from the global rules; see the methodology changelog.

October 13, 2026: OMP moves to the Agent Plugins package

The @maestria/omp native package is retired. OMP coverage continues through @maestria/agent-plugins, which supplies seven task profiles, the /fein, /sonar, and /blitz commands, an always-apply rule, and four skills (maestria install omp installs it). OMP owns activation and task execution; the package declares no runtime extension or task wrapper.
Retired with the package: session-start agent deployment, before_agent_start mode injection, compaction and session-state preservation, goal mirroring, tool-call interception and review-mode enforcement, the maestria_subagent wrapper, the executable’s /fein, /sonar, /blitz, /review, /restore-model, /review-model, /handoff, and /maestria-status commands (the plugin supplies its own /fein, /sonar, and /blitz), and per-agent model files (maestria configure omp).
Migrating: run omp plugin uninstall @maestria/omp (and remove any maestria agent files left in ~/.omp/agent/agents/), then maestria install omp. Use the /fein, /sonar, and /blitz commands, and OMP’s task interface when delegating.

v0.8.0 (Pi) / v0.6.0 (OMP)

October 5, 2026

Pi 1.0 and the Oh My Pi 18.x line are supported hosts. @maestria/pi accepts Pi ^1.0.0 and @maestria/omp admits Oh My Pi 18.x; no adapter behavior change on either side.
Fresh Pi installs no longer hit a subagent peer conflict. The @gotgenes/pi-subagents peer moved to its supported 21.x range, matching what the documented install path already installs.
The Node.js floor for the Pi-family packages is now 22.19+, matching the Pi host requirement.
Pi specialist agent files no longer name a glob tool that Pi does not register; Oh My Pi output, where glob is a real builtin, is unchanged.

v0.7.18 (Pi)

October 1, 2026

Subagent polling and parallel result handling were simplified; timeout, abort, and sibling cleanup behavior are preserved.

v0.7.17 (Pi) / v0.5.17 (OMP)

September 30, 2026

Agents reassess explanatory comments and redundant tests before accepting changes, and shared directive text was consolidated with no behavior change; see the methodology changelog.

v0.7.16 (Pi) / v0.5.16 (OMP)

September 25, 2026

Agents follow behavior-first testing guidance and deliver work as verified, reviewable slices with designs shown before implementation; see the methodology changelog.
OMP only: the opt-in methodology annotation hook and mode footer middleware were removed, and mode-switch pipeline text is unified in one table.

v0.7.15 (Pi) / v0.5.15 (OMP)

September 24, 2026

Project customization (.maestria/workflow.md, .maestria/rules.md) now loads through one shared loader across integrations; ordering, live reloads, and error reporting are unchanged.

v0.7.14 (Pi) / v0.5.14 (OMP)

September 23, 2026

Project customization (.maestria/workflow.md then .maestria/rules.md) loads as subordinate guidance; see the methodology changelog.
create-pull-request and docs-update skills install through the skills CLI with per-skill selection; see the CLI changelog.
Delegation wording clarified as contract-driven delegation; docs only, no behavior change. See the methodology changelog.

v0.7.13 (Pi) / v0.5.13 (OMP)

September 18, 2026

Consistent PR delivery contract with proportional documentation assessment; see the methodology changelog.

v0.7.12 (Pi) / v0.5.12 (OMP)

September 17, 2026

Stricter visual delivery evidence contract, with capture, PR-body publication, and readback as distinct stages; see the methodology changelog.

v0.7.11 (Pi) / v0.5.11 (OMP)

September 16, 2026

Defined when Sonar research is complete, with Blitz reusing valid context and switching routes for investigation or design work; see the methodology changelog.

v0.7.10 (Pi) / v0.5.10 (OMP)

September 15, 2026

Clarified rendered verification and visual delivery evidence, keeping required artifacts and verification gaps in handoffs; see the methodology changelog.

v0.7.9 (Pi) / v0.5.9 (OMP)

September 15, 2026

Internal consolidation; no user-facing behavior change and generated agent and skill output is unchanged.

v0.7.8 (Pi) / v0.5.8 (OMP)

September 11, 2026

Internal consolidation; no user-facing behavior change and generated agent and skill output is unchanged.

v0.7.7 (Pi) / v0.5.7 (OMP)

September 8, 2026

Stopped publishing generated JavaScript sourcemaps to reduce published package sizes; compiled runtime artifacts are unchanged.

v0.7.6 (Pi) / v0.5.6 (OMP)

September 6, 2026

Cross-cutting agent methodology update; see the methodology changelog.

v0.7.5 (Pi) / v0.5.5 (OMP)

September 4, 2026

Cross-cutting agent methodology update; see the methodology changelog.

v0.7.4 (Pi) / v0.5.4 (OMP)

September 3, 2026

Cross-cutting agent methodology update; see the methodology changelog.

v0.7.3 (Pi) / v0.5.3 (OMP)

August 27, 2026

Shared neutral mode and skill validation logic through the hybrid package topology.

v0.7.2 (Pi) / v0.5.2 (OMP)

August 25, 2026

Cross-cutting agent methodology update; see the methodology changelog.

v0.7.1 (Pi) / v0.5.1 (OMP)

August 24, 2026

Cross-cutting agent methodology update; see the methodology changelog.

v0.7.0 (Pi) / v0.5.0 (OMP)

August 22, 2026

Cross-cutting agent methodology update; see the methodology changelog.

v0.6.8 (Pi)

August 13, 2026

Fix chain-mode {previous} substitution so previous results containing $ sequences are inserted literally, and abort already-spawned subagents when a later parallel spawn fails so none are orphaned.

v0.6.7 (Pi)

August 13, 2026

Effect v4 polling

Use Effect v4 RC for cancellable subagent polling and structured parallel dispatch, so aborted or failed polls no longer leave sibling polling work running (historical Effect adoption).

v0.6.6 (Pi) / v0.4.3 (OMP)

August 13, 2026

Pi: parallel subagent dispatch preserves successful results and aborts orphaned subagents on a partial failure.
Cross-cutting agent methodology update; see the methodology changelog.

v0.5.6 (Pi) / v0.2.0 (OMP)

July 21-22, 2026

Cross-cutting agent methodology update; see the methodology changelog.

v0.5.5 (Pi)

July 21, 2026

Compact agent directives

Cross-cutting agent methodology update; see the methodology changelog.

v0.5.4 (Pi)

July 14, 2026

Cross-cutting agent methodology update; see the methodology changelog.

v0.5.3 (Pi)

July 14, 2026

Mandatory PR description structure

Cross-cutting agent methodology update; see the methodology changelog.

v0.5.2 (Pi)

July 14, 2026

Cross-cutting agent methodology update; see the methodology changelog.

v0.5.1 (Pi)

July 14, 2026

Cross-cutting agent methodology update; see the methodology changelog.

v0.4.1 (Pi)

July 11, 2026

Standard Pi/OMP skill mechanism for agent content

The orchestrator dispatcher prompt and global agent rules now load as platform skills (.md files in skills/). The old custom file-loading approach is removed.

Specialist agents registered with subagent system

All 7 specialist agent types (adventurer, architect, builder, diagnose, planner, reviewer, writer) are now registered as proper agent types. Each specialist has the right tool permissions - read-only specialists (adventurer, architect, planner, reviewer, diagnose) can’t write files; builder and writer can.

v0.4.0 (Pi)

July 10, 2026

Cross-cutting agent methodology update; see the methodology changelog.

v0.3.7 (Pi)

July 1, 2026

Code intelligence tool preference

Cross-cutting agent methodology update; see the methodology changelog.

v0.3.1 (Pi)

June 27, 2026

Two new design principles

Cross-cutting agent methodology update; see the methodology changelog.

v0.3.0 (Pi)

June 24, 2026

Cross-cutting agent methodology update; see the methodology changelog.

v0.2.1 (Pi)

June 24, 2026

Single source of truth for agent directives

Cross-cutting agent methodology update; see the methodology changelog.

v0.2.0 - Initial release

June 23-24, 2026 (Pi) / July 22, 2026 (OMP)

Features

Initial release of @maestria/pi and @maestria/omp - portable extensions that bring maestria’s spec-driven orchestration, workflow modes, and specialist subagents to the Pi and OMP coding agents.

7 specialist agent types - adventurer, architect, builder, diagnose, planner, reviewer, writer with structured role definitions and tool permission boundaries
3 workflow modes - fein (full pipeline with gates), sonar (research only), blitz (fast implementation), each routing to the appropriate subagent pipeline
Global rules injection - cross-cutting orchestration, delegation, and context-management rules injected into every session
Session compaction preservation - structured state summaries survive compaction events, keeping task tracking and work-in-progress intact across turns
Subagent dispatch - via @gotgenes/pi-subagents (Pi) or native task() tool (OMP), with 6-field handoff validation
/orchestrate, /review, /maestria-status commands - slash commands for dispatching work, requesting reviews, and inspecting extension state
Maker/checker enforcement - tool-level restriction prevents the orchestrator from implementing its own tasks
4 methodology skills - orchestrator dispatcher, global agent rules, handoff contract, and iteration limits

OMP-specific

Built-in task() dispatch - uses OMP’s native task() tool, no separate subagent package needed

v0.1.0 (Pi)

June 22, 2026

Features

Initial release of @maestria/pi - a portable Pi extension bringing spec-driven orchestration, workflow modes, and specialist subagents to the Pi coding agent.

8 specialist prompt templates - one orchestrator plus seven specialist subagents
3 workflow modes - fein, sonar, blitz
Global rules injection via before_agent_start event
Session compaction preservation
Subagent dispatch via @gotgenes/pi-subagents with 6-field handoff validation
/orchestrate, /review, /maestria-status commands
Maker/checker enforcement via tool-call interception
2 methodology skills - handoff contract and iteration limits