A safety-first tool for performing high-risk maintenance on Pinax vaults, including diagnostics, organization, and version restoration.
Install
mkdir -p .claude/skills/pinax-proof-maintenance-operator && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/16841" && unzip -o skill.zip -d .claude/skills/pinax-proof-maintenance-operator && rm skill.zipInstalls to .claude/skills/pinax-proof-maintenance-operator
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Use when an agent needs to run Pinax proof loops, vault doctor/stats, metadata/repair/organize plans or applies, version snapshot/restore, record ledger maintenance, or other high-risk local maintenance with snapshot and approval gates.Key capabilities
- →Run Pinax proof loops
- →Perform vault doctor and stats operations
- →Manage metadata plans and applies
- →Execute repair and organize plans
- →Create and restore version snapshots
- →Conduct record ledger maintenance
How it works
This skill executes Pinax commands for diagnostics, planning, and applying changes, ensuring safety through read-only checks, plan inspections, and pre-maintenance snapshots.
Inputs & outputs
When to use pinax-proof-maintenance-operator
- →Running vault diagnostics
- →Applying organization metadata plans
- →Creating pre-maintenance version snapshots
- →Restoring vault state from history
About this skill
Pinax Proof Maintenance Operator
Run Pinax proof-loop and high-risk maintenance workflows with explicit diagnostics, saved plans, snapshots, and approval gates. This skill owns maintenance that can modify many notes, .pinax/** state, version evidence, or recovery state.
Use When
- The request mentions
pinax proof, proof loop, doctor/stats, maintenance, metadata completion, repair, organize, version snapshot, restore, rollback, record ledger, or broad cleanup. - The user wants to apply a saved repair/organize/metadata plan.
- A command may rewrite frontmatter, move many notes, repair indexes/metadata, restore vault state, or mutate record/version evidence.
Command Patterns
pinax proof loop run --vault ./my-notes --json
pinax proof loop run --vault ./my-notes --apply --yes --json
pinax vault doctor --vault ./my-notes --json
pinax vault stats --vault ./my-notes --agent
pinax metadata plan --vault ./my-notes --json
pinax metadata apply --vault ./my-notes --yes --json
pinax repair plan --vault ./my-notes --save --json
pinax repair apply --vault ./my-notes --plan repair-abc123 --yes --snapshot-message "pre-repair snapshot" --json
pinax organize plan --vault ./my-notes --save --json
pinax organize list --vault ./my-notes --agent
pinax organize apply --vault ./my-notes --plan organize-abc123 --yes --snapshot-message "pre-organization snapshot" --json
pinax version snapshot --vault ./my-notes --message "before maintenance" --json
pinax version restore notes/a.md --revision snap_123 --plan --vault ./my-notes --json
pinax version restore apply --vault ./my-notes --plan restore-abc123 --yes --json
pinax record status --vault ./my-notes --agent
pinax record adopt --vault ./my-notes --plan --json
Workflow
- Inspect first with
pinax vault doctor --json,pinax vault stats --agent, orpinax proof loop run --json. - Prefer a read-only proof loop before manual sequencing.
pinax proof loop run --jsoncovers capture/retrieve/diagnose/plan/snapshot/apply readiness in one bounded projection. - Generate and inspect plans before any apply:
pinax metadata plan,pinax repair plan --save, orpinax organize plan --save. - Before
metadata apply,repair apply,organize apply,version restore, broad record adoption, or any high-risk write, create a fresh snapshot withpinax version snapshot --message "before maintenance" --json. - Require explicit user approval before commands with
--yes,--apply, restore, destructive cleanup, or bulk moves. - After an apply, run the matching diagnostic again, such as
pinax vault doctor --json,pinax proof loop run --json, orpinax record status --agent. - If an operation reports partial success, conflict, snapshot required, or approval required, stop and surface the exact next safe command.
Safety Boundaries
- Do not hand-edit
.pinax/**, SQLite stores, repair/organize plan JSON, record ledgers, or version metadata. - Do not run
--yes,--apply, or restore commands without a snapshot and explicit approval. - Do not treat
proof loop run --apply --yesas a default command; use the read-only proof loop first. - Do not hide warnings about untracked files, missing snapshots, conflicts, or unsupported restore backends.
Validation
- Before apply: saved plan path exists in command output and a fresh snapshot command succeeded.
- After apply:
pinax proof loop run --jsonorpinax vault doctor --jsonreports expected status. - For restore: run a read-only status/doctor command and report restored snapshot id, not raw file diffs.
When not to use it
- →When direct editing of `.pinax/**` files or SQLite stores is required
- →When running `--yes`, `--apply`, or restore commands without prior snapshot and explicit approval
- →When treating `proof loop run --apply --yes` as a default command without a read-only proof loop first
Limitations
- →Does not support hand-editing `.pinax/**` files, SQLite stores, or plan JSON
- →Requires explicit user approval for destructive operations
- →Does not hide warnings about untracked files, missing snapshots, or conflicts
How it compares
This workflow provides explicit diagnostic steps, saved plans, snapshots, and approval gates, unlike manual command execution.
Compared to similar skills
pinax-proof-maintenance-operator side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| pinax-proof-maintenance-operator (this skill) | 0 | 1mo | Review | Intermediate |
| interactive-shell | 0 | 1mo | No flags | Intermediate |
| project-management | 0 | 3mo | No flags | Beginner |
| resolve-conflicts | 81 | 8mo | Review | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
interactive-shell
Jonghakseo
dev server, TUI, REPL, DB shell, 로그처럼 사용자 제어나 장시간 실행이 필요한 터미널 작업에 사용한다. AI 작업 위임에는 subagent를 사용한다.
project-management
elhaddajiOtmane
Automated workflows for Git version control, database backups, semantic naming, and task tracking. Use this skill whenever the user asks to save progress, commit changes, or perform project management tasks.
resolve-conflicts
antinomyhq
Use this skill immediately when the user mentions merge conflicts that need to be resolved. Do not attempt to resolve conflicts directly - invoke this skill first. This skill specializes in providing a structured framework for merging imports, tests, lock files (regeneration), configuration files, and handling deleted-but-modified files with backup and analysis.
openspec-onboard
studyzy
Guided onboarding for OpenSpec - walk through a complete workflow cycle with narration and real codebase work.
codex-cli-bridge
alirezarezvani
Bridge between Claude Code and OpenAI Codex CLI - generates AGENTS.md from CLAUDE.md, provides Codex CLI execution helpers, and enables seamless interoperability between both tools
skill-sync
KyleKing
Syncs Claude Skills with other AI coding tools like Cursor, Copilot, and Codeium by creating cross-references and shared knowledge bases. Invoke when user wants to leverage skills across multiple tools or create unified AI context.