Skill Lifecycle Governance Skill
Authoritative skill lifecycle state model to prevent invalid states or transitions.
The skill describes permission boundaries and governance actions, but as a knowledge document, it does not execute permissions. It clearly lists who can perform which actions and emphasizes review processes and audit logs, but lacks details on user confirmation or data flow transparency. Deduction for lacking explicit disclosure of data flow and sensitive data handling.
The document provides clear state transitions and code references, with specific descriptions of key paths (publish, review, withdraw). However, no tests or failure examples are included, and static review cannot verify execution, so deduction.
The skill clearly defines trigger conditions and applicable scenarios, but lacks non-fit range boundaries. Environment fit: description does not depend on overseas services, so no deduction.
Documentation is well-structured with state model, transitions, permission boundaries, and common pitfalls, but lacks install/dependency notes, versioning and changelog. Deduction for missing maintenance responsibility and update path.
The skill serves as a reference document that can answer lifecycle-related questions, but direct usability is limited as it provides no concrete outputs or examples. Deduction for unclear deliverable.
The skill references specific code files and design docs, but provides no independent verification material or test evidence, so evidence strength is limited.
- This skill is a knowledge document only, with no executable code; manual reference to the codebase is required for actual use.
- Publisher is unverified; independent verification of source and update support is recommended.
- The skill references internal code and design docs but provides no tests or independent validation; static review cannot confirm accuracy.
What does this skill do, and when should you use it?
This skill provides the complete state model for skill containers, versions, review tasks, and visibility overlays, along with all valid state transitions and governance actions (hide, yank, archive). It is based on the SkillHub codebase and design docs, explicitly noting discrepancies between design intent and code (e.g., the hidden flag preferred over an enum state). As a reference-only skill, it performs no operations; it guides agents in modifying publish, review, or unpublish flows to follow the correct state machine and permission boundaries.
The skill defines authoritative enums (SkillStatus, SkillVersionStatus, ReviewTaskStatus, SkillVisibility), lists all valid transitions with triggers (first upload, review approve/reject, withdraw, yank, hide, archive, etc.), specifies the latestVersionId pointer recalculation rules, explains read-model projections (headlineVersion, publishedVersion, etc.) and permission boundaries (who can withdraw, delete, archive, hide, etc.). It also records domain events (SkillPublishedEvent, SkillStatusChangedEvent) and common pitfalls (setting SkillStatus.HIDDEN directly, forgetting to recalc latestVersionId, etc.).
- A developer modifying the publish flow needs to verify that new state transitions conform to the defined legal transition table.
- A backend engineer adding or changing status fields consults the canonical enum definitions and design notes.
- An engineer implementing governance actions (hide, yank, archive) checks the correct implementation and permissions.
- When displaying skill state in search or detail pages, a developer uses the skill to understand which projection to use (publishedVersion, ownerPreviewVersion).
- A code reviewer validates new state transitions or permission checks against the skill's rules.
What are this skill's strengths and limitations?
- Provides an authoritative state model, reducing design inconsistencies
- Explicitly flags design-vs-code discrepancies, guiding new code
- Detailed transition tables, permission boundaries, and common pitfalls are practical
- Documentation-only, easy to integrate into any agent workflow
- Specific to SkillHub; not applicable to other registries
- No executable scripts or tests; reference only
- Depends on specific class names and docs that may drift from the codebase
- Does not cover all edge cases (e.g., concurrent publishing)
How do you install this skill?
This is a documentation-only skill without executable files. To use it, place the .agents/skills/skill-lifecycle folder in your agent's skills directory. If using the SkillHub registry, install via CLI (e.g., skillhub install skill-lifecycle --agent codex).
How do you use this skill?
Prompt your agent to reference this skill when modifying skill publish, review, or unpublish flows. For example: 'Consult the skill-lifecycle skill to ensure the new state transitions are valid.' The agent will use the state model and transition tables to validate your code or design.
How does this skill compare with similar options?
None. The skill is based on SkillHub's internal design and does not mention other similar tools.