feat: add first-class Oh My Pi support by camopy · Pull Request #256 · compozy/compozy

camopy · 2026-07-20T23:09:04Z

Summary

register Oh My Pi (omp) as a distinct native ACP runtime without changing Pi
preserve OMP-owned model selection while applying Compozy reasoning and permission semantics
install skills into OMP project and active-profile roots across setup, preflight, and doctor flows
document the runtime and cover its catalog, execution, resume, setup, and public CLI contracts

Why

Oh My Pi exposes a native ACP server and profile-aware skill roots that differ from the existing Pi adapter. First-class support lets users run Compozy workflows through omp acp while preserving OMP's active model selection and leaving all existing Pi behavior unchanged.

Testing

make verify
5,000 Go tests passed; 8 gated/helper tests skipped
5 Playwright smoke tests passed
Go lint completed with 0 issues
manually executed an workflow task end to end with --ide omp --model auto --reasoning-effort medium

Summary by CodeRabbit

New Features
- Added Oh My Pi (OMP) support across IDE selection, setup/installation/verification, and supported-agent discovery (including listings).
- Enabled OMP execution with session resumption, tool-call routing, and permission handling, plus live end-to-end smoke coverage.
Bug Fixes
- Improved bundled-skill preflight/refresh behavior for interactive flows and missing skill handling.
- Refined model/reasoning configuration resolution and OMP session setup validation.
- Improved permission “allow-once” selection behavior and added clearer failure handling for missing OMP runtime and unsupported reasoning effort.

compozyrabbit · 2026-07-20T23:09:12Z

Code Review Could Not Complete ⚠️

The review failed before suggestions could be generated.

Reason: The configured API key (openai) is out of credits or has hit its billing limit. Top up the account or adjust the plan.

After fixing the issue, comment @kody review on this PR to re-run the review.

Kody Guide: Usage and Configuration

Interacting with Kody

Request a Review: Ask Kody to review your PR manually by adding a comment with the @kody start-review command at the root of your PR.
Validate Business Logic: Ask Kody to validate your code against business rules by adding a comment with the @kody -v business-logic command.
Provide Feedback: Help Kody learn and improve by reacting to its comments with a 👍 for helpful suggestions or a 👎 if improvements are needed.

Current Kody Configuration

Review Options

The following review options are enabled or disabled:

Options	Enabled
Bug	✅
Performance	✅
Security	✅
Business Logic	✅

Access your configuration settings here.

coderabbitai · 2026-07-20T23:09:41Z

Walkthrough

This PR adds Oh My Pi as a supported runtime, including public identifiers, registry metadata, environment and installation resolution, ACP session configuration, CLI setup and execution flows, skill preflight handling, and persisted-session coverage.

Changes

Oh My Pi runtime integration

Layer / File(s)	Summary
Runtime contracts and catalog registration `internal/core/model/constants.go`, `internal/core/agent/registry_specs.go`, `internal/core/api.go`, `compozy.go`, `internal/cli/form.go`, `internal/core/agent/tool_call_name.go`	Adds Oh My Pi identifiers, default model metadata, `omp acp` registry commands, catalog entries, CLI guidance, and common ACP tool aliases.
Environment resolution and setup installation `internal/setup/*`, `internal/cli/setup.go`, `internal/cli/extension/doctor.go`	Resolves OMP and PI profiles/configuration directories, exposes OMP as a setup agent, lists supported agents, and verifies installations.
ACP session configuration and client behavior `internal/core/agent/session_config.go`, `internal/core/agent/client.go`, `internal/core/agent/*_test.go`	Propagates refreshed ACP options and covers OMP model defaults, reasoning validation, permissions, resume behavior, and live sessions.
CLI execution and skill preflight `internal/cli/root_command_execution_test.go`, `internal/cli/skills_preflight.go`, `internal/cli/skills_preflight_test.go`	Adds OMP CLI execution and setup journeys, profile-aware bundled-skill refreshes, interactive missing-skill handling, and resolver snapshot reuse.
Persisted execution and ACP helper harness `internal/core/run/exec/*`, `internal/core/run/executor/execution_acp_integration_test.go`	Tests OMP ACP arguments, workspace CWD propagation, streamed tool updates, setup failures, and persisted session loading.

Estimated code review effort: 5 (Critical) | ~120 minutes

Possibly related PRs

compozy/compozy#60: Provides the exec integration test foundation extended with OMP execution and resume scenarios.
compozy/compozy#70: Overlaps with the runtime registry and catalog infrastructure used for the OMP specification.

Suggested reviewers: pedronauck

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 1.60% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly matches the main change: adding first-class Oh My Pi support.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands.}

greptile-apps · 2026-07-20T23:11:15Z

Greptile Summary

This PR adds first-class Oh My Pi support throughout Compozy. The main changes are:

Registers omp acp as a distinct runtime without changing Pi.
Preserves OMP model selection while applying reasoning and permission settings.
Adds profile-aware OMP skill setup, preflight, verification, and doctor support.
Exposes OMP through the CLI and public Go API.
Adds runtime, setup, resume, and integration test coverage.
Documents OMP installation and execution behavior.

Confidence Score: 5/5

This looks safe to merge.

The updated setup flow preserves OMP profile and destination settings across verification and refresh.
Interactive runs remain blocked when required bundled skills are missing and refresh is declined.
Runtime coverage includes execution, resume, configuration, permissions, and missing-runtime behavior.
No blocking issue remains within the scope of this follow-up review.

Important Files Changed

Filename	Overview
internal/core/agent/registry_specs.go	Registers OMP as a separate native ACP runtime with its own command and capability settings.
internal/core/agent/session_config.go	Adds OMP model and reasoning configuration for new and resumed sessions.
internal/setup/agents.go	Adds project and active-profile skill roots for OMP while keeping Pi destinations separate.
internal/cli/skills_preflight.go	Keeps skill verification, refresh, and re-verification on the same resolved destination.
internal/cli/setup.go	Adds environment-aware OMP setup resolution and supported-agent listing output.

_{Reviews (2): Last reviewed commit: "docs: document Oh My Pi support" | Re-trigger Greptile}

coderabbitai

Actionable comments posted: 1

🧹 Nitpick comments (2)

internal/setup/agents_test.go (1)
179-181: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick win

Subtest names should use the Should... pattern. The new table/loop subtests label cases as "Traversal", "Absolute path", the profile string, and tt.name (e.g. "Default profile uses native root"). Renaming to the mandated t.Run("Should ...") form keeps these enforced-standard-compliant.

As per path instructions: "MUST use t.Run("Should...") pattern for ALL test cases".

Also applies to: 282-283, 356-358
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@internal/setup/agents_test.go` around lines 179 - 181, Rename every subtest
case name in the affected tests to use the mandated “Should...” pattern,
including the loop labels “Traversal” and “Absolute path,” the profile-string
cases, and tt.name-based cases. Update the corresponding t.Run calls near the
referenced test sections while preserving each case’s existing behavior.
Source: Path instructions
internal/core/agent/session_config_test.go (1)
44-44: 📐 Maintainability & Code Quality | 🔵 Trivial | 💤 Low value

Subtest names don't follow the mandated Should… pattern.

The subtest cases here ("empty model", "explicit auto") use free-form names. As per path instructions, test cases must use the t.Run("Should...") pattern for all test cases. This applies to the new OMP tests across this cohort.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@internal/core/agent/session_config_test.go` at line 44, Update the subtest
names in the t.Run calls within the affected OMP tests to follow the mandated
“Should…” naming pattern, including the cases currently named “empty model” and
“explicit auto”; preserve each test’s existing behavior and assertions.
Source: Path instructions

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@internal/core/agent/registry_test.go`:
- Around line 193-219: Enforce the required Should... test naming across all
affected sites: wrap the top-level test logic in t.Run subtests named with the
Should prefix in internal/core/agent/registry_test.go ranges 193-219, 1199-1230,
and 1616-1669, internal/cli/form_test.go ranges 56-91 and 794-821, and
test/public_api_test.go range 16-31; rename table-driven name fields in
internal/core/agent/registry_test.go range 411-427 and
internal/core/agent/tool_call_name_test.go range 77-83 to start with Should; and
prefix the dynamic t.Run name in internal/core/agent/registry_test.go range
449-460 with Should.

---

Nitpick comments:
In `@internal/core/agent/session_config_test.go`:
- Line 44: Update the subtest names in the t.Run calls within the affected OMP
tests to follow the mandated “Should…” naming pattern, including the cases
currently named “empty model” and “explicit auto”; preserve each test’s existing
behavior and assertions.

In `@internal/setup/agents_test.go`:
- Around line 179-181: Rename every subtest case name in the affected tests to
use the mandated “Should...” pattern, including the loop labels “Traversal” and
“Absolute path,” the profile-string cases, and tt.name-based cases. Update the
corresponding t.Run calls near the referenced test sections while preserving
each case’s existing behavior.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

Push a commit to this branch (recommended)
Create a new PR with the fixes

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: e0ab0955-1d22-434a-8b51-6882aa8070b1

📥 Commits

Reviewing files that changed from the base of the PR and between 00dba6a and da0dde0.

⛔ Files ignored due to path filters (4)

README.md is excluded by !**/*.md
skills/compozy/SKILL.md is excluded by !**/*.md
skills/compozy/references/cli-reference.md is excluded by !**/*.md
skills/compozy/references/config-reference.md is excluded by !**/*.md

📒 Files selected for processing (30)

compozy.go
internal/cli/extension/doctor.go
internal/cli/extension/doctor_test.go
internal/cli/form.go
internal/cli/form_test.go
internal/cli/root_command_execution_test.go
internal/cli/setup.go
internal/cli/setup_test.go
internal/cli/skills_preflight.go
internal/cli/skills_preflight_test.go
internal/core/agent/client_test.go
internal/core/agent/omp_live_test.go
internal/core/agent/registry_specs.go
internal/core/agent/registry_test.go
internal/core/agent/session_config.go
internal/core/agent/session_config_test.go
internal/core/agent/tool_call_name.go
internal/core/agent/tool_call_name_test.go
internal/core/api.go
internal/core/model/constants.go
internal/core/run/exec/exec_integration_test.go
internal/core/run/exec/test_helpers_test.go
internal/core/run/executor/execution_acp_integration_test.go
internal/setup/agents.go
internal/setup/agents_test.go
internal/setup/install_test.go
internal/setup/runtime_agents.go
internal/setup/types.go
internal/setup/verify_test.go
test/public_api_test.go

compozyrabbit · 2026-07-21T00:32:02Z

Code Review Could Not Complete ⚠️

The review failed before suggestions could be generated.

Reason: The configured API key (openai) is out of credits or has hit its billing limit. Top up the account or adjust the plan.

After fixing the issue, comment @kody review on this PR to re-run the review.

Kody Guide: Usage and Configuration

Interacting with Kody

Request a Review: Ask Kody to review your PR manually by adding a comment with the @kody start-review command at the root of your PR.
Validate Business Logic: Ask Kody to validate your code against business rules by adding a comment with the @kody -v business-logic command.
Provide Feedback: Help Kody learn and improve by reacting to its comments with a 👍 for helpful suggestions or a 👎 if improvements are needed.

Current Kody Configuration

Review Options

The following review options are enabled or disabled:

Options	Enabled
Bug	✅
Performance	✅
Security	✅
Business Logic	✅

Access your configuration settings here.

coderabbitai

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (1)

internal/cli/root_command_execution_test.go (1)
1-1: 📐 Maintainability & Code Quality | 🟠 Major | ⚡ Quick win

Wrap new OMP tests in t.Run("Should...") subtests.

All five new OMP test functions run their assertions directly in the top-level Test... function body instead of a t.Run("Should...", ...) subtest, which the path instructions for **/*_test.go require for every test case.

internal/cli/root_command_execution_test.go#L140-202: wrap TestExecCommandOMPCompletesThroughPublicDaemonPath's body in t.Run("Should complete exec through the public daemon path", func(t *testing.T) {...}).

internal/cli/root_command_execution_test.go#L204-227: wrap TestExecCommandOMPMissingRuntimeReportsNativeInstallGuidance's body in a t.Run("Should report native install guidance when omp is missing", ...) subtest.

internal/cli/root_command_execution_test.go#L229-266: wrap TestExecCommandOMPUnsupportedReasoningReportsAdvertisedChoices's body in a t.Run("Should report advertised reasoning choices when unsupported", ...) subtest.

internal/cli/root_command_execution_test.go#L268-310: wrap TestSetupCommandOMPProjectCopyJourney's body in a t.Run("Should install project skills via copy journey", ...) subtest.

internal/cli/root_command_execution_test.go#L312-381: wrap TestSetupCommandOMPNamedProfileGlobalCopyJourney's body in a t.Run("Should install global named-profile skills via copy journey", ...) subtest.

As per path instructions, "MUST use t.Run("Should...") pattern for ALL test cases."
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@internal/cli/root_command_execution_test.go` at line 1, Wrap the assertion
bodies of TestExecCommandOMPCompletesThroughPublicDaemonPath,
TestExecCommandOMPMissingRuntimeReportsNativeInstallGuidance,
TestExecCommandOMPUnsupportedReasoningReportsAdvertisedChoices,
TestSetupCommandOMPProjectCopyJourney, and
TestSetupCommandOMPNamedProfileGlobalCopyJourney in t.Run subtests using the
specified “Should...” descriptions, passing the subtest *testing.T to each body.
Preserve the existing test setup, assertions, and behavior.
Source: Path instructions

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Outside diff comments:
In `@internal/cli/root_command_execution_test.go`:
- Line 1: Wrap the assertion bodies of
TestExecCommandOMPCompletesThroughPublicDaemonPath,
TestExecCommandOMPMissingRuntimeReportsNativeInstallGuidance,
TestExecCommandOMPUnsupportedReasoningReportsAdvertisedChoices,
TestSetupCommandOMPProjectCopyJourney, and
TestSetupCommandOMPNamedProfileGlobalCopyJourney in t.Run subtests using the
specified “Should...” descriptions, passing the subtest *testing.T to each body.
Preserve the existing test setup, assertions, and behavior.

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 9f2a1b0c-bc86-4954-bb55-f5c2a332ef6e

📥 Commits

Reviewing files that changed from the base of the PR and between da0dde0 and c9f3f52.

⛔ Files ignored due to path filters (4)

README.md is excluded by !**/*.md
skills/compozy/SKILL.md is excluded by !**/*.md
skills/compozy/references/cli-reference.md is excluded by !**/*.md
skills/compozy/references/config-reference.md is excluded by !**/*.md

📒 Files selected for processing (33)

compozy.go
internal/cli/extension/doctor.go
internal/cli/extension/doctor_test.go
internal/cli/form.go
internal/cli/form_test.go
internal/cli/root_command_execution_test.go
internal/cli/setup.go
internal/cli/setup_test.go
internal/cli/skills_preflight.go
internal/cli/skills_preflight_test.go
internal/core/agent/client.go
internal/core/agent/client_test.go
internal/core/agent/omp_live_test.go
internal/core/agent/registry_specs.go
internal/core/agent/registry_test.go
internal/core/agent/session_config.go
internal/core/agent/session_config_test.go
internal/core/agent/terminal_lifecycle_test.go
internal/core/agent/tool_call_name.go
internal/core/agent/tool_call_name_test.go
internal/core/api.go
internal/core/model/constants.go
internal/core/run/exec/exec_integration_test.go
internal/core/run/exec/test_helpers_test.go
internal/core/run/executor/execution_acp_integration_test.go
internal/setup/agents.go
internal/setup/agents_test.go
internal/setup/install_test.go
internal/setup/legacy_cleanup.go
internal/setup/runtime_agents.go
internal/setup/types.go
internal/setup/verify_test.go
test/public_api_test.go

🚧 Files skipped from review as they are similar to previous changes (26)

internal/setup/runtime_agents.go
compozy.go
internal/core/agent/tool_call_name.go
internal/core/api.go
internal/core/model/constants.go
internal/core/agent/registry_specs.go
internal/cli/form.go
internal/cli/form_test.go
internal/core/agent/tool_call_name_test.go
internal/cli/skills_preflight_test.go
internal/core/run/exec/exec_integration_test.go
internal/core/agent/omp_live_test.go
internal/setup/agents.go
internal/setup/types.go
internal/setup/verify_test.go
internal/cli/skills_preflight.go
internal/core/agent/client_test.go
internal/cli/setup.go
test/public_api_test.go
internal/core/run/executor/execution_acp_integration_test.go
internal/cli/extension/doctor_test.go
internal/cli/extension/doctor.go
internal/core/run/exec/test_helpers_test.go
internal/cli/setup_test.go
internal/core/agent/registry_test.go
internal/core/agent/session_config.go

pedronauck · 2026-07-28T02:53:20Z

Closing as part of the CompozyOS v0.3 reset.

Compozy was rebuilt from the ground up as CompozyOS — an agent operating system. main was replaced by the v0.3 rewrite, and the first beta (v0.3.0-beta.1) is landing through the beta release channel — so this diff no longer applies to the current codebase. The 0.2.x line this PR targets is preserved on the legacy/v0.2 branch and only receives critical patches.

If you would like this capability in CompozyOS, please reopen it against the new architecture — happy to help route it.

coderabbitai Bot requested changes Jul 20, 2026

View reviewed changes

Comment thread internal/core/agent/registry_test.go

camopy added 5 commits July 20, 2026 21:28

feat: register omp runtime foundation

587735d

feat: implement omp acp execution

877a442

feat: add native omp skill roots

df6de55

feat: wire omp setup journeys

d1e320d

docs: document Oh My Pi support

c9f3f52

camopy force-pushed the feat/omp branch from da0dde0 to c9f3f52 Compare July 21, 2026 00:31

coderabbitai Bot reviewed Jul 21, 2026

View reviewed changes

coderabbitai Bot approved these changes Jul 21, 2026

View reviewed changes

coderabbitai Bot mentioned this pull request Jul 25, 2026

fix: resolve inherited cross-runtime models against runtime defaults #259

Merged

pedronauck closed this Jul 28, 2026

This was referenced Aug 12, 2026

fix: resolve extension-published agents for session command catalogs #350

Merged

refactor: unify PRD and TechSpec into a single spec pipeline #397

Merged

Navigation Menu

Repository navigation

feat: add first-class Oh My Pi support - #256

feat: add first-class Oh My Pi support#256
camopy wants to merge 5 commits into
compozy:mainfrom
camopy:feat/omp

camopy commented Jul 20, 2026 •

edited by coderabbitai Bot

Loading

Uh oh!

compozyrabbit Bot commented Jul 20, 2026 •

edited

Loading

Uh oh!

coderabbitai Bot commented Jul 20, 2026 •

edited

Loading

❌ Failed checks (1 warning)

Uh oh!

greptile-apps Bot commented Jul 20, 2026 •

edited

Loading

Greptile Summary

Uh oh!

coderabbitai Bot left a comment

Uh oh!

Uh oh!

compozyrabbit Bot commented Jul 21, 2026 •

edited

Loading

Uh oh!

coderabbitai Bot left a comment

Uh oh!

pedronauck commented Jul 28, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

camopy commented Jul 20, 2026 • edited by coderabbitai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Why

Testing

Summary by CodeRabbit

Uh oh!

compozyrabbit Bot commented Jul 20, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Code Review Could Not Complete ⚠️

Uh oh!

coderabbitai Bot commented Jul 20, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Walkthrough

Changes

❌ Failed checks (1 warning)

Uh oh!

greptile-apps Bot commented Jul 20, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Greptile Summary

Confidence Score: 5/5

Important Files Changed

Uh oh!

coderabbitai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

compozyrabbit Bot commented Jul 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Code Review Could Not Complete ⚠️

Uh oh!

coderabbitai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

pedronauck commented Jul 28, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

camopy commented Jul 20, 2026 •

edited by coderabbitai Bot

Loading

compozyrabbit Bot commented Jul 20, 2026 •

edited

Loading

coderabbitai Bot commented Jul 20, 2026 •

edited

Loading

greptile-apps Bot commented Jul 20, 2026 •

edited

Loading

compozyrabbit Bot commented Jul 21, 2026 •

edited

Loading