npx claudepluginhub terrylica/cc-skills --plugin itpThis skill is limited to using the following tools:
references/evolution-log.mdSearches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.
Guides idea refinement into designs: explores context, asks questions one-by-one, proposes approaches, presents sections for approval, writes/review specs before coding.
Delegate to the repository's mise release tasks. Every repo should define its own release DAG in .mise/tasks/release/.
Self-Evolving Skill: This skill improves through use. If instructions are wrong, parameters drifted, or a workaround was needed — fix this file immediately, don't defer. Only update for real, reproducible issues.
# Check if the repo has mise release tasks
mise tasks ls 2>/dev/null | grep -i release
mise run release:full (or the repo's equivalent)| Flag | Short | Description |
|---|---|---|
--dry | -d | Dry-run mode (preview, no modifications) |
--status | -s | Show current version and release state |
# Full release (preflight → version → sync → verify)
mise run release:full
# Dry-run (preview what would be released)
mise run release:dry
# Check current state
mise run release:status
| Concern | Prescriptive Skill | mise Task Delegation |
|---|---|---|
| Repo-specific logic | Duplicated in skill | Lives in .mise/tasks/ |
| DAG enforcement | Manual ordering | depends array enforced |
| Maintainability | Update skill + tasks | Single source in tasks |
| Portability | Assumes npm/bun | Uses whatever the repo configures |
| Secrets | Hardcoded patterns | [env] in .mise.toml |
Repos should follow the hub-and-spoke pattern:
.mise.toml # Hub: [env] + [tools] + task docs
.mise/tasks/
└── release/
├── _default # Help / navigation
├── preflight # Phase 1: Validate prerequisites
├── version # Phase 2: Bump version (semantic-release)
├── sync # Phase 3: Sync artifacts (marketplace, cache)
├── verify # Phase 4: Verify release artifacts
├── full # Orchestrator: depends on all phases
├── dry # Dry-run preview
└── status # Current version info
┌──────────┐
│ preflight│
└─────┬────┘
│ depends
┌─────▼────┐
│ version │
└─────┬────┘
│ sequential
┌─────▼────┐
│ sync │
└─────┬────┘
│ sequential
┌─────▼────┐
│ verify │
└──────────┘
# .mise/tasks/release/full (orchestrator)
depends = ["release:preflight"]
# Chains: preflight → version → sync → verify
# .mise/tasks/release/preflight (guard)
# Checks: clean dir, auth, plugins, releasable commits
# .mise/tasks/release/version (core)
depends = ["release:preflight"]
# Runs: semantic-release (or language-specific versioning)
If the repo has no mise release tasks, read the semantic-release skill:
Read: ${CLAUDE_PLUGIN_ROOT}/skills/semantic-release/SKILL.md
Read: ${CLAUDE_PLUGIN_ROOT}/skills/semantic-release/references/local-release-workflow.md
Then follow the 4-phase workflow documented there.
| Error | Resolution |
|---|---|
mise tasks not found | Install mise: curl https://mise.run | sh |
| No release tasks | Create .mise/tasks/release/ or use fallback skill |
| Working dir not clean | git stash or commit changes |
| Not on main branch | git checkout main |
| No releasable commits | Create a feat: or fix: commit first |
| Wrong account | Check GH_TOKEN / GH_ACCOUNT in .mise.toml [env] |
After this skill completes, check before closing:
Only update if the issue is real and reproducible — not speculative.