Legacy Per-Host Changelog

These entries preserve the release history of the standalone @maestria/claude-code, @maestria/codex, @maestria/cursor, @maestria/kimi-code, @maestria/hermes, and @maestria/omp packages. All of them were consolidated into @maestria/agent-plugins in October 2026; see the Agent Plugins changelog for current releases and migration notes. Methodology changes shared across hosts are tracked in the methodology changelog.

Early smoke checks recorded against the pre-split archive layouts (Claude Code 2.1.217 offline fixture, OMP 18.4.8 packed discovery, Hermes isolated-home smoke, Codex CLI 0.145.0 marketplace smoke) predate the native resource split and do not validate the current layout; see compatibility for current evidence.

Claude Code

Former package: @maestria/claude-code.

v0.3.13

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.

v0.3.12

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.

v0.3.11

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.

v0.3.10

September 18, 2026

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

v0.3.9

September 17, 2026

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

v0.3.8

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.

v0.3.7

September 15, 2026

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

v0.3.6

September 11, 2026

• Internal consolidation; duplicated specialist guidance was removed from generated orchestrator skills, with the guidance unchanged in substance.

v0.3.5

September 6, 2026

• Clarified installation paths, prerequisites, and support boundaries in the package documentation.
• Cross-cutting agent methodology update; see the methodology changelog.

v0.3.4

September 4, 2026

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

v0.3.3

September 3, 2026

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

v0.3.2

August 25, 2026

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

v0.3.1

August 24, 2026

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

v0.3.0

August 22, 2026

Leaner directives, explicit delivery autonomy

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

v0.2.1

Marketplace and CLI installation

• The published package can be installed, checked, updated, and removed with maestria install claude-code, maestria status, maestria update claude-code, and maestria uninstall claude-code.
• The maestria CLI stages the npm package in a local marketplace and delegates installation state to Claude Code’s native plugin manager.
• Exact version pinning is not supported for maestria update claude-code --version; the host marketplace selects the latest package.
• Plugin metadata now stays aligned with package versions: release-time version sync is generalized across plugins and a --check mode fails the build on drift, so the published plugin artifact can no longer diverge from npm metadata.
• Cross-cutting agent methodology update; see the methodology changelog.

v0.2.0

Initial release

Declarative Claude Code plugin

@maestria/claude-code is a declarative Claude Code plugin generated from the canonical directives in packages/core/agent-directives/:

• 7 specialist agents with global-rules skill preloading - adventurer, architect, builder, diagnose, planner, reviewer, writer - available as @maestria:<agent> mentions
• Orchestrator and global-rules skills - /maestria:orchestrator routes work to specialists; maestria:global-rules is preloaded into every agent
• Workflow commands - /maestria:fein, /maestria:sonar, and /maestria:blitz select the pipeline depth
• Read-only enforcement - adventurer, planner, and reviewer deny Write/Edit via disallowedTools (user-authorized; the only runtime enforcement)
• Local validation - claude plugin validate --strict is the automated gate; marketplace installation is available through Claude Code or the maestria CLI

Shared directive updates

The plugin tracks the shared agent directives as they evolve. Cross-cutting changes reflected in this release are tracked in the methodology changelog.

Codex

Former package: @maestria/codex.

0.4.11

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.

0.4.10

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.

0.4.9

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.

0.4.8

September 18, 2026

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

0.4.7

September 17, 2026

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

0.4.6

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.

0.4.5

September 15, 2026

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

0.4.4

September 11, 2026

• Internal consolidation; duplicated specialist guidance was removed from generated orchestrator skills, with the guidance unchanged in substance.

0.4.3

September 6, 2026

• Clarified installation paths, prerequisites, and support boundaries in the package documentation.
• Cross-cutting agent methodology update; see the methodology changelog.

0.4.2

September 4, 2026

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

0.4.1

September 3, 2026

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

0.4.0

August 27, 2026

Native Codex orchestration

• The maestria CLI now installs seven native maestria-* Codex custom-agent roles.
• The CLI adds a marked global instruction block so Codex’s host-owned primary session activates maestria orchestration and delegates with native agent_type role names.
• Updates preserve user model, reasoning, and service-tier settings, and uninstall removes only maestria-managed agent files and instruction content.

0.3.2

August 25, 2026

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

0.3.1

August 24, 2026

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

0.3.0

August 22, 2026

Leaner directives, explicit delivery autonomy

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

0.2.1

August 13, 2026

maestria CLI support

• Added maestria CLI compatibility: the CLI detects the Codex CLI host, stages the published npm package in a local marketplace, and delegates install, update, status, check, and uninstall operations to Codex’s native plugin manager.

0.2.0

Codex CLI projection

• Added the generated skills projection for the 7 specialists, orchestrator, global rules, handoff, iteration limits, and fein/sonar/blitz workflow modes.
• Added maestria CLI compatibility: install, status, check, update, and uninstall now recognize the codex platform.
• The CLI stages the npm package in a local Codex marketplace and delegates plugin state to Codex.
• Codex desktop parity, hooks, MCP, model configuration, and hard read-only enforcement remain out of scope.

0.1.0

Initial Codex CLI projection. It packages the canonical maestria specialist, orchestration, rules, handoff, iteration-limit, and workflow-mode directives as Codex skills through .codex-plugin/plugin.json.

Cursor

Former package: @maestria/cursor.

0.2.14

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.

0.2.13

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.

0.2.12

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.

0.2.11

September 18, 2026

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

0.2.10

September 17, 2026

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

0.2.9

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.

0.2.8

September 15, 2026

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

0.2.7

September 11, 2026

• Internal consolidation; duplicated specialist guidance was removed from generated orchestrator skills, with the guidance unchanged in substance.

0.2.6

September 6, 2026

• Clarified installation paths, prerequisites, and support boundaries in the package documentation.
• Cross-cutting agent methodology update; see the methodology changelog.

0.2.5

September 4, 2026

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

0.2.4

September 3, 2026

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

0.2.3

August 27, 2026

• Aligned Cursor packaging and native per-agent model configuration while preserving existing user settings across plugin updates.

0.2.2

August 25, 2026

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

0.2.1

August 24, 2026

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

0.2.0

August 22, 2026

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

0.1.8

August 13, 2026

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

0.1.7

August 12-13, 2026

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

0.1.6

August 11, 2026

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

0.1.5

August 6, 2026

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

0.1.4

August 3, 2026

Selective routing contract

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

0.1.1

July 21, 2026

Minor Changes

• Initial release of @maestria/cursor - declarative Cursor plugin for IDE and CLI.
• 7 specialist agents synced from core (agents/*.md) with Cursor-adapted tool names
• Orchestrator skill (skills/orchestrator/SKILL.md) with Task-based routing, handoff contracts, and maker/checker enforcement
• Global rules (rules/maestria-global.mdc, alwaysApply: true)
• Workflow commands - /fein (full pipeline), /sonar (research only), /blitz (fast implementation)
• Two-layer maker/checker - readonly: true runtime flag on adventurer/planner/reviewer agents
• CLI support - maestria install cursor, maestria update cursor, maestria uninstall cursor, maestria check cursor
• Cross-cutting agent methodology update; see the methodology changelog.

0.1.0

July 21, 2026

Initial release - same as 0.1.0 (versioning alignment).

Kimi Code

Former package: @maestria/kimi-code.

v0.6.11

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.

v0.6.10

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.

v0.6.9

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.

v0.6.8

September 18, 2026

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

v0.6.7

September 17, 2026

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

v0.6.6

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.

v0.6.5

September 15, 2026

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

v0.6.4

September 11, 2026

• Internal consolidation; duplicated specialist guidance was removed from generated orchestrator skills, with the guidance unchanged in substance.

v0.6.3

September 6, 2026

• Removed the duplicated Kimi orchestrator skill prescription so sync.config.ts only adapts toward the plugin runtime.
• Cross-cutting agent methodology update; see the methodology changelog.

v0.6.2

September 4, 2026

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

v0.6.1

September 3, 2026

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

v0.6.0

August 27, 2026

• Updated the Kimi Code integration to use its native plugin registry, system-prompt delivery, namespaced commands, and current delegation contracts.

v0.5.2

August 25, 2026

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

v0.5.1

August 24, 2026

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

v0.5.0

August 22, 2026

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

v0.4.16

August 13, 2026

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

v0.4.15

August 12-13, 2026

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

v0.4.14

August 11, 2026

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

v0.4.13

August 6, 2026

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

v0.4.12

August 3, 2026

Selective routing contract

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

v0.4.11

July 29, 2026

Rule-override conditions and communication conventions

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

v0.4.10

July 28, 2026

Blind review and fail-loud iteration exit

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

v0.4.9

July 22, 2026

Directive recomposition

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

v0.4.8

July 21, 2026

• Dependency updates across packages (diff, zod, @clack, effect, astro, and more). No user-facing behavior change.

v0.4.7

July 21, 2026

Switch to npm-based install

The plugin now installs via npm pack @maestria/kimi-code instead of git-based codeload. maestria update kimi-code no longer re-downloads the same version.

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

v0.4.6

July 21, 2026

Compact agent directives

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

v0.4.5

July 14, 2026

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

v0.4.4

July 14, 2026

Structured PR descriptions

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

v0.4.3

July 14, 2026

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

v0.4.2

July 14, 2026

Structured work results are now mandatory

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

v0.4.1

July 10, 2026

• Reordered conventional commit prefixes. refactor is now the default prefix instead of feat. feat is reserved for user-facing features only.

v0.4.0

July 10, 2026

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

v0.3.5

July 1, 2026

Code intelligence tool preference

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

v0.3.1

June 27, 2026

Two new design principles

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

v0.3.0

June 24, 2026

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

v0.2.1

June 24, 2026

Single source of truth for agent directives

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

0.1.0

Minor Changes

• Initial release of @maestria/kimi-code - a declarative, manifest-based plugin that ports the 8 specialist skills from @maestria/opencode v0.3.2 to Kimi Code v0.13.1.
• 8 SKILL.md files - orchestrator + 7 specialists using the v0.3.x Skill Prescription pattern
• Manifest-only plugin - no SDK hooks, no build step; the plugin is pure static config (kimi.plugin.json + skills/ + rules/)
• Swarm-aware orchestration - the orchestrator routes ≥3 uniform items through Kimi Code’s AgentSwarm for parallel fan-out, with the 7-specialist table covering the coder / explore / plan subagent profiles
• Global rules - rules/AGENTS.md with the 7-specialist delegation table, auto-loaded by Kimi Code from ~/.kimi-code/AGENTS.md
• Install checklist - INSTALL.md with recommended [[hooks]] and [[permission.rules]] config blocks for tool-layer safety enforcement (persona text alone is advisory)
• 20 vitest tests - validate manifest schema, frontmatter, rules budget, and persona safety constraints (reviewer no-edit, adventurer read-only Bash)

See ADR-KC-001 for the full design rationale.

Hermes

Former package: @maestria/hermes.

• Replaced the Python adapter with portable skills in @maestria/agent-plugins. Native mode commands, gating, hooks, and the OpenCode bridge are retired. Entries below preserve the former adapter release history.

Former Python Adapter (unreleased)

• Project customization (.maestria/workflow.md then .maestria/rules.md) loads as subordinate guidance, re-read every pre_llm_call turn with a visible error banner; see the methodology changelog.

0.1.23 (2026-10-01)

• /review and /plan now reach Hermes’ built-in commands instead of being swallowed by the plugin’s mode switch. The plugin registers /fein, /sonar, /blitz, /mode, and /mode-clear; its reviewer and planner specialists run as /fein pipeline stages.
• Sonar mode can now read installed skills and recall past sessions (skill_view, skills_list, session_search), matching what delegated children already could. skill_manage stays blocked, and no write, shell, code-execution, or delegation tool was added to any mode.

0.1.22 (2026-09-30)

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

0.1.21 (2026-09-25)

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

0.1.20 (2026-09-18)

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

0.1.19 (2026-09-17)

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

0.1.18 (2026-09-16)

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

0.1.17 (2026-09-11)

• Removed legacy role and permission scaffolding, unified slash-command mode text, and restored the fail-closed denial for unknown mode values; generated skill guidance no longer describes the removed role model.

0.1.16 (2026-09-06)

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

0.1.15 (2026-09-04)

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

0.1.14 (2026-09-03)

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

0.1.13 (2026-08-12)

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

0.1.12 (2026-08-11)

• Cross-cutting agent methodology update; see the methodology changelog.
• Hardened the Hermes adapter by wiring native session and subagent lifecycle trust, immutable mode and child safety allowlists, and the current Hermes plugin registration contracts.

0.1.11 (2026-08-04)

Fixed

• Fixed the Hermes delegation rules adaptation in the synced global rules. The route-scoped rewrite of the canonical Delegation section had left a sync findAndReplace transform silently no-oping, so the generated global-rules/SKILL.md shipped the raw canonical wording instead of the Hermes adaptation. Hermes agents again get the correct guidance: on focused/full routes, use only the 7 maestria specialists and never delegate to Hermes’ built-in explore or general agents, which bypass the pipeline.

0.1.10 (2026-08-03)

Selective routing contract

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

0.1.9 (2026-07-29)

Rule-override conditions and communication conventions

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

0.1.8 (2026-07-28)

Blind review and fail-loud iteration exit

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

0.1.7 (2026-07-22)

Directive recomposition

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

0.1.6 (2026-07-21)

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

0.1.5 (2026-07-21)

Compact agent directives

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

0.1.4 (2026-07-17)

Changed

• Orchestrator behavior: Simple tasks now run directly instead of always being delegated to subagents. Complex tasks (multi-file, cross-domain, risky) still use specialist routing. Routine changes are faster with less overhead - no action needed on your part.

0.1.3 (2026-07-17)

Added

• pre_gateway_dispatch hook: Intercepts /fein, /sonar, /blitz, /mode, /review, /plan commands before the agent-busy check
• Role-based permission enforcement (historical implementation): [MAESTRIA_ROLE: <specialist>] in delegate_task context enforced per-specialist tool restrictions
• Transform hook: Write operations in fein/blitz mode append methodology annotation to tool results

Fixed

• Telegram forum topic responses now route to the correct thread instead of General

0.1.2 (2026-07-16)

Fixed

• Plugin loading failure due to src/ layout - added root __init__.py shim

0.1.1 (2026-07-16)

Initial release - same as 0.1.0 (versioning alignment).

0.1.0 (2026-07-16)

Added

• Plugin scaffold with plugin.yaml + register(ctx) entry point
• 9 skills: 7 specialists + orchestrator + global rules
• Mode system with file persistence (/fein, /sonar, /blitz, /mode)
• pre_llm_call hook for mode injection into user message
• pre_tool_call hook for mode-based tool gating (sonar write-block)
• llm_execution middleware (opt-in mode footer via MAESTRIA_MODE_FOOTER=1)
• 7 slash commands: /fein, /sonar, /blitz, /mode, /mode-clear, /review, /plan
• OpenCode CLI routing tool (opencode_route) for delegating complex coding tasks
• Subagent lifecycle hooks (subagent_start/stop for pipeline tracking)
• Session lifecycle hooks (on_session_start/end)
• transform_tool_result hook (observer-mode write logging)
• Distribution via hermes plugins install from this repository (git-based)

Changed (post-release design doc alignment)

• Memory engine agnosticism - removed stale Mnemosyne integration claim, removed memory/kanban probes
• Dead code removed: memory.py (MemoryManager.record() never called)
• Cross-platform detection: shutil.which() replaces subprocess.run(["which", ...])
• Plugin check: node -e "require(...)" replaces fragile npm filesystem walker

Oh My Pi

Former package: @maestria/omp.

2026-10-10

• Moved OMP to @maestria/agent-plugins and grouped its public guides under /agent-plugins/omp/. The initial 14-skill, seven-profile layout is superseded by the native resource split described in the Agent Plugins changelog.