Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Wiki Maintainer

ASecurity

Use when the user asks to mine recent sessions for recurring agent failures, write durable diagnostic patterns, maintain the global skill-evolution wiki, or bootstrap a recurring Scheduled Task so Wiki Maintainer keeps running without further commands. Triggers on "update the skill wiki", "record this failure pattern", "analyze my sessions for skill gaps", "알아서 돌려", "매일 wiki", "자동으로 정리", or "wiki-maintainer".

10 stars
0 votes
0 copies
0 views
Added 9/28/2026
ai-agentspythongoshellbash

Works with

cli

Security Analysis

A100/100

Scanned 9/28/2026

Install to Claude Code

$npx -y skills add fritzprix/libr-agent --skill wiki-maintainer --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Wiki Maintainer?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Wiki Maintainer
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/fritzprix-wiki-maintainer/badge)](https://www.skillsdirectory.com/skills/fritzprix-wiki-maintainer)

More formats (shields.io, HTML) on the badges page.

Files
SKILL.md
---
name: wiki-maintainer
description: >
  Use when the user asks to mine recent sessions for recurring agent failures,
  write durable diagnostic patterns, maintain the global skill-evolution wiki,
  or bootstrap a recurring Scheduled Task so Wiki Maintainer keeps running
  without further commands. Triggers on "update the skill wiki", "record this
  failure pattern", "analyze my sessions for skill gaps", "알아서 돌려",
  "매일 wiki", "자동으로 정리", or "wiki-maintainer".
---

# Wiki Maintainer

Turn **past session history** into durable patterns in the **host-global** wiki
so **later sessions on this machine** can reuse them (Windows, macOS, and Linux).
Meta workflow only: do not inject wiki files into a normal task agent. Skill
patches go through **skill-proposer**.

**Preferred UX:** one bootstrap chat → create a global Scheduled Task that wakes
**Wiki Maintainer** on a cron. See [autonomous-loop.md](references/autonomous-loop.md).

## Prerequisites

- Prefer the bundled **Wiki Maintainer** assistant (ships with `history` + workspace shell).
- Otherwise enable optional **`history`** plus shell/code execution for this skill’s Python CLI.
- Tool names use `server__tool` form. Match the live tool list.
- **Wiki root:** `Path.home() / ".libragent" / "wiki"`  
  - Linux/macOS: `~/.libragent/wiki`  
  - Windows: `%USERPROFILE%\.libragent\wiki`
- Resolve this skill’s Base Directory as `<skill-base-dir>`.
- Recurring automation uses core **`scheduled_task__*`** via the **schedule** skill.

### Invoking the CLI (all OS)

Prefer `python` on PATH; on Linux/macOS fall back to `python3` if needed:

```bash
python "<skill-base-dir>/scripts/wiki_cli.py" path
python "<skill-base-dir>/scripts/wiki_cli.py" init
```

Windows PowerShell: same commands; quoting with `"..."` is fine. For large Unicode
pattern bodies, write a UTF-8 file with `workspace__writeFile` then:

```bash
python "<skill-base-dir>/scripts/wiki_cli.py" write-pattern --id <kebab-id> --file <utf8-path>
```

Templates: [wiki-layout.md](references/wiki-layout.md),
[index-template.md](references/index-template.md),
[logs-template.md](references/logs-template.md),
[skill-impact-template.md](references/skill-impact-template.md),
[pattern-template.md](references/pattern-template.md),
[analysis-rubric.md](references/analysis-rubric.md),
[autonomous-loop.md](references/autonomous-loop.md).

## Not this skill

| Need | Use |
| --- | --- |
| Facts / preferences into the knowledge graph | **knowledge-distiller** |
| Propose a concrete `SKILL.md` patch | **skill-proposer** |
| Repo doc wiki (`[[slug]]`, catalog.json) | **repo-wiki** |
| One-shot clock delay inside this chat only | **loop** |
| Wait for process/ticket/webhook completion | **call-me-back** |
| Create/update the recurring wake | **schedule** (see autonomous-loop) |

## Workflow

### 0. Bootstrap automation (when the user wants “알아서”)

Follow [autonomous-loop.md](references/autonomous-loop.md): init wiki → optional
first mine → `agent__listAgents` for Wiki Maintainer id →
`scheduled_task__createScheduledTask` with the wake prompt. Then stop asking
the user to re-run this skill manually.

### 1. Ensure global wiki exists

```bash
python "<skill-base-dir>/scripts/wiki_cli.py" init
python "<skill-base-dir>/scripts/wiki_cli.py" path
```

Never use `{workspace}/.libragent/wiki` for this evolution store.

### 2. Choose sessions

1. Scope: today, last N sessions, or named ids.
2. `history__listSessions` / `history__searchHistory`.
3. Prefer failures, long recovery loops, user corrections. Skip chit-chat.
4. On scheduled wakes, skip prior Wiki Maintainer / skill-proposer meta runs.

### 3. Analyze

1. `history__readSession`, then `history__readMessage` as needed.
2. Capture symptom, first divergence, owning layer ([analysis-rubric.md](references/analysis-rubric.md)).
3. Seek a successful counterexample in another session.
4. One session only → `status: draft`.

### 4. Write or update a pattern

```bash
python "<skill-base-dir>/scripts/wiki_cli.py" write-pattern --id <kebab-id> --file <utf8-path>
python "<skill-base-dir>/scripts/wiki_cli.py" upsert-index --id <kebab-id> --status draft --one-line "<summary>"
python "<skill-base-dir>/scripts/wiki_cli.py" prepend-log --message "<what you found>"
python "<skill-base-dir>/scripts/wiki_cli.py" cat patterns/<kebab-id>.md
```

Structure: [pattern-template.md](references/pattern-template.md).

### 5. Stop conditions

- Do not edit `SKILL.md` here — hand off to **skill-proposer**.
- On scheduled runs, skill-proposer may **Propose** only; never Accept without
  an interactive user Accept.
- No evaluation-task answers or hidden grader hints.
- Only record what `history__*` returned.

## Safety

- De-dupe with `list-patterns` / `cat index.md`.
- Host-global wiki may retain data across projects — do not store secrets.
- Do not create duplicate scheduled tasks for the same purpose.

Attribution

fritzprixfritzprix
View sourceMore from fritzprix →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1074701 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

695601 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

691 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →