Version comparison
Two versions pinned to commits, side by side. Differences are marked as facts; no side is declared better.
Swap sides Change left Change right
Package names, types, authors, licenses and repository details are current registry metadata, not their values at the selected commits; historical values for these fields were not recorded. Components are pinned to commits. Dependencies and compatibility show only plugin.json or marketplace-entry fields saved with these snapshots (manifest fields). Declarations in a skill's SKILL.md header are separately collected, later evidence: they take no part in this comparison and never rewrite these fields; see “Skill file declarations” on a single version page.
| Property | ||
|---|---|---|
| Package Differs | outputai | superpowers |
| Current package type | Plugin | Plugin |
| Repository Differs | growthxai/output | obra/superpowers |
| Package path Differs | coding_assistants/claude/plugins/outputai |
Repository root |
| Current author | Output.ai | Jesse Vincent |
| Current license | Apache-2.0 | MIT |
| Version label Differs | 0.3.0 | 6.4.2 |
| Commit Differs | a4f6bd40ab0e3a51f524c1dc07524f33fe6999ad |
8ca22dba9a94f28898bbce59f2537ff4d87c747d |
| Captured Differs | ||
| Manifest file | Manifest | Manifest |
| Total components Differs | 57 | 16 |
| Skill Differs | 51 | 15 |
| Agent Differs | 5 | 0 |
| Hook Same | 1 | 1 |
| Recorded dependency count Same | 0 | 0 |
| Recorded compatibility | Not recorded | Not recorded |
| Marketplace Differs | claude-plugins-official anthropics/claude-plugins-official | superpowers-dev obra/superpowers |
| Collection warnings Same | 0 | 0 |
| Repository stars | 440 | 293,533 |
| Download requests (package) | 0 | 1 |
| Repo updated | ||
| Source archive | Download | Download |
Component comparison
Matched by component type and name. This compares names and paths only, not content; source-file evidence follows in the next section.
Only in outputai 0.3.0 (a4f6bd4)56
- Agent workflow_context_fetcher
- Agent workflow_debugger
- Agent workflow_planner
- Agent workflow_prompt_writer
- Agent workflow_quality
- Skill output-build-workflow
- Skill output-credentials-edit
- Skill output-credentials-env-vars
- Skill output-credentials-init
- Skill output-debug-workflow
- Skill output-dev-agent-class
- Skill output-dev-code-style
- Skill output-dev-cost-hooks
- Skill output-dev-create-skeleton
- Skill output-dev-credentials
- Skill output-dev-eval-testing
- Skill output-dev-evaluator-function
- Skill output-dev-folder-structure
- Skill output-dev-http-client-create
- Skill output-dev-llm-streaming
- Skill output-dev-model-selection
- Skill output-dev-prompt-file
- Skill output-dev-scenario-file
- Skill output-dev-skill-file
- Skill output-dev-step-function
- Skill output-dev-types-file
- Skill output-dev-upgrade-prompt-models
- Skill output-dev-workflow-cost
- Skill output-dev-workflow-function
- Skill output-error-direct-io
- Skill output-error-http-client
- Skill output-error-missing-schemas
- Skill output-error-nondeterminism
- Skill output-error-try-catch
- Skill output-error-zod-import
- Skill output-eval-audit
- Skill output-eval-dataset-design
- Skill output-eval-error-analysis
- Skill output-eval-judge-prompt
- Skill output-eval-validate-judge
- Skill output-meta-post-flight
- Skill output-meta-pre-flight
- Skill output-meta-project-context
- Skill output-migrate
- Skill output-plan-workflow
- Skill output-services-check
- Skill output-workflow-list
- Skill output-workflow-reset
- Skill output-workflow-result
- Skill output-workflow-run
- Skill output-workflow-runs-list
- Skill output-workflow-start
- Skill output-workflow-status
- Skill output-workflow-stop
- Skill output-workflow-trace
- Skill output-workflow-trace-file
Only in superpowers 6.4.2 (8ca22db)15
- Skill brainstorming
- Skill diagnosing-superpowers
- Skill dispatching-parallel-agents
- Skill executing-plans
- Skill finishing-a-development-branch
- Skill receiving-code-review
- Skill requesting-code-review
- Skill subagent-driven-development
- Skill systematic-debugging
- Skill test-driven-development
- Skill using-git-worktrees
- Skill using-superpowers
- Skill verification-before-completion
- Skill writing-plans
- Skill writing-skills
Same name, different path1
- Hook SessionStart
coding_assistants/claude/plugins/outputai/hooks/hooks.jsonhooks/hooks.json
In both1
- Hook SessionStartPath differs
Source-file evidence
Compares source-file bytes and modes only. Missing evidence is not counted as unchanged.
Evidence scope and limits
Compares only the one file each component's source link points to at its pinned commit (Git blob and file mode), not a whole skill directory: SKILL.md stands for that one file. When several components share one manifest, hooks or settings file, a change to that file does not show that every component in it changed. This is not a semantic, quality or behavior comparison.
Source-file comparison is limited to revisions of the same package. Component names, paths and other records remain comparable across packages; neither side is ranked.
Recorded dependency comparison
No dependency requirements were recorded from plugin.json or marketplace entries. Only manifest fields saved with the snapshot are shown here; dependency declarations in other files are outside the current collection scope.
Collection warnings
Left
No collection warnings.
Right
No collection warnings.
No winner is declared. A comparison lists what is recorded for two pinned versions. Stars are a repository-level signal and download requests are requests this site received; the registry does not collect token usage, cache hits or benchmarks, so none appear here. Metric definitions