refactor(artifact): extract info operation

[mnriem]

Summary

• extract the typed, transport-independent artifact.info logical operation beneath specify artifact info
• refactor the existing CLI command into a thin adapter while preserving its exact JSON, stream, error, and exit-code behavior
• add positive, negative, parity, isolation, Unicode, hook, layering, and error-conversion coverage

Architecture

The new operation uses explicit project context and does not depend on process cwd, SPECIFY_INIT_DIR, Typer, Rich, MCP, CLI consoles, subprocesses, or os.chdir().

• operation ID: artifact.info
• contract version: 1
• capability: local-read
• network access: none
• request: explicit absolute project directory, artifact identifier, and optional established artifact kind
• result: typed named-artifact or hook-artifact data preserving the complete ordered composition stack
• expected failures: project path/project validation, invalid kind or identifier, unknown/ambiguous artifacts, unknown hooks, resolution/filesystem failures, and invalid catalog results

Pure request validation, including identifier grammar, runs before state-dependent Spec Kit project validation. This deliberately does not add or register an MCP adapter. The artifact MCP inventory keeps artifact.info unavailable while sourcing its contract version, capabilities, and network metadata from the shared descriptor.

Validation

• .venv/bin/python -m pytest -q tests/specify_cli/artifacts/test_operation_info.py tests/specify_cli/artifacts/test_command_info.py --cov=specify_cli.artifacts._operation_info --cov=specify_cli.artifacts.command_info --cov-report=term-missing --cov-branch — 69 passed; 100% statements and 100% branches
• LC_ALL=en_US.UTF-8 .venv/bin/python -m pytest -q tests/specify_cli/artifacts tests/specify_cli/test_command_mcp.py tests/specify_cli/test_mcp_version.py — 288 passed
• LC_ALL=en_US.UTF-8 .venv/bin/python -m pytest -q — 10,136 passed, 19 skipped
• uvx ruff check src/specify_cli/artifacts/_operation_info.py tests/specify_cli/artifacts/test_operation_info.py — passed
• uvx ruff format --check src/specify_cli/artifacts/_operation_info.py tests/specify_cli/artifacts/test_operation_info.py — passed
• git diff --check — passed
• full pytest collection increased from 10,087 to 10,155 after rebasing onto current upstream/main

Scope audit

No changes were made to artifact list pagination or MCP response budgets, artifacts/mcp_list.py, artifacts/_operation_list.py, artifact lookup, MCP tool registration, or JSON contract normalization. The only MCP inventory change keeps artifact.info unavailable while aligning its static metadata and reason with the new shared descriptor.

AI disclosure

This pull request and its review fixes were implemented and validated autonomously by GitHub Copilot using GPT-5.6 Sol on behalf of @mnriem. The maintainer explicitly approved opening this fourth concurrent pull request after being informed that repository policy may assign it lower review priority.

[Copilot]

🟡 Changes recommended

Error classification and MCP inventory metadata remain inconsistent with the new operation contract.

3 open findings

• Honor explicit kind over conflicting hook shorthand · New
• Map all structurally invalid IDs to invalid_artifact_identifier · New
• Align MCP inventory metadata with the shared artifact descriptor · New

What changed in this PR

Extracts artifact.info into a typed, transport-independent operation while retaining the CLI adapter.

Changes:

• Adds typed request, result, descriptor, and error contracts.
• Refactors the CLI command to dispatch through the shared operation.
• Adds operation and CLI parity/error coverage.

File	Description
src/​specify_cli/​artifacts/​_operation_info.py	Implements the shared operation and contracts.
src/​specify_cli/​artifacts/​command_info.py	Converts the CLI into a thin adapter.
tests/​specify_cli/​artifacts/​test_operation_info.py	Tests operation behavior and validation.
tests/​specify_cli/​artifacts/​test_command_info.py	Tests CLI mapping and output compatibility.

🧠 Review effort: Balanced

---

Give feedback about Copilot approvals in this survey to enter a drawing for a $150 gift card.

[mnriem]

Addressed all findings from review 5457087760 in c4c6aa3667e359cbd649ae5f7c08d9da0d2bbe51:

• explicit non-hook kinds now take precedence over conflicting hook: shorthand when classifying not-found errors
• accepted artifact identifier grammar is validated before catalog lookup, including unknown prefixes and malformed named/hook IDs, while explicit hook event:command identifiers remain supported
• the unavailable artifact.info MCP inventory entry now sources contract version, capabilities, and network metadata from the shared descriptor and states that the adapter is not yet implemented

Validation after rebasing onto current upstream/main: 68 focused tests with 100% statement/branch coverage, 287 artifact/MCP regressions, and the full suite at 10,135 passed / 19 skipped. Ruff, formatting, and git diff --check pass.

AI disclosure: this review round was implemented, validated, and summarized autonomously by GitHub Copilot using GPT-5.6 Sol on behalf of @mnriem.

[Copilot]

🟡 Changes recommended

Structural identifier validation must precede state-dependent project validation to follow the shared operation lifecycle.

1 open finding

• Validate artifact identifiers before project-state checks · New

3 resolved since last review

• Map all structurally invalid IDs to invalid_artifact_identifier
• Honor explicit kind over conflicting hook shorthand
• Align MCP inventory metadata with the shared artifact descriptor

🧠 Review effort: Balanced

---

Give feedback about Copilot approvals in this survey to enter a drawing for a $150 gift card.

[mnriem]

Addressed review 5457561822 in a10794f4699b39ff9323bc92903a9b4f89b2a811 by moving structural identifier validation ahead of the .specify project-state check. A malformed identifier now deterministically returns invalid_artifact_identifier even when the selected directory is not a Spec Kit project. Added a regression proving the catalog is not reached and pure request validation takes precedence.

Validation: 69 focused tests with 100% statement/branch coverage, 288 artifact/MCP regressions, and the full suite at 10,136 passed / 19 skipped. Ruff, formatting, and git diff --check pass.

AI disclosure: this review round was implemented, validated, and summarized autonomously by GitHub Copilot using GPT-5.6 Sol on behalf of @mnriem.

[Copilot]

🟢 Approval recommended

The implementation follows the shared-operation architecture and includes thorough positive, negative, parity, and isolation coverage.

0 open findings

1 resolved since last review

• Validate artifact identifiers before project-state checks

🧠 Review effort: Balanced

---

Give feedback about Copilot approvals in this survey to enter a drawing for a $150 gift card.