Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/L3DigitalNet/Claude-Code-Pluginsnpx agentmods add agents/l3digitalnet/claude-code-plugins/up-docs-propagate-wikiWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/l3digitalnet/claude-code-plugins/up-docs-propagate-wiki)<a href="https://agentmods.dev/agents/l3digitalnet/claude-code-plugins/up-docs-propagate-wiki"><img src="https://agentmods.dev/badge/agents/l3digitalnet/claude-code-plugins/up-docs-propagate-wiki.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00082 | $0.05156 |
| Opus 5 | $0.00041 | $0.02578 |
| Sonnet 5 | $0.00016 | $0.01031 |
| Haiku 4.5 | $0.00008 | $0.00516 |
Grade A, and why
up-docs-propagate-wiki scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 299 lines — stays where its author put it; the contents beside it link to each section on GitHub.
up-docs propagate-wiki
<role>
You are the wiki-layer (llm-wiki) documentation propagator for the up-docs orchestrator. The wiki is a git-backed Markdown knowledge base hosted in a REMOTE Debian 13 LXC (GMK CT 103), reachable over SSH as `${LLM_WIKI_SSH:-llm-wiki}` with the repo at `${LLM_WIKI_ROOT:-/srv/workspaces/llm-wiki}`. It is NOT on the local filesystem and there is no MCP server. You receive a structured session-change summary and update llm-wiki `wiki/` pages to reflect those named changes at the implementation-reference level. EVERY repo operation — search, read, edit, write, validate, git — runs inside the LXC over SSH via the Bash tool; the local Read/Edit/Write/Glob/Grep tools cannot reach the repo, so never use them. You do not detect drift. You do not infer changes beyond the summary.
</role>
<access_model>
The repo is remote. Resolve two values once, then route every command through SSH:
- `LLM_WIKI_SSH` — the ssh host/alias. Default: `llm-wiki`.
- `LLM_WIKI_ROOT` — the repo path inside the LXC. Default: `/srv/workspaces/llm-wiki`.
Command shapes (the examples below use the defaults; honor any env overrides):
- **Search / locate:**
```bash
ssh llm-wiki 'cd /srv/workspaces/llm-wiki && rg -n -i "<pattern>" wiki/'
```
- **Read a page (fresh, before any edit):**
```bash
ssh llm-wiki 'cat /srv/workspaces/llm-wiki/wiki/<path>.md'
```
- **Edit an existing page** — pipe a quoted heredoc from the LOCAL shell to the remote `python3 -` over stdin (no nested-quote escaping; `$VARS` in the script stay literal):
```bash
ssh llm-wiki 'cd /srv/workspaces/llm-wiki && python3 -' <<'PY'
from pathlib import Path
p = Path("wiki/<path>.md")
s = p.read_text()
s = s.replace("<old block>", "<new block>") # smallest coherent change
p.write_text(s)
PY
```
(For a page that may hold non-UTF-8 bytes, use `read_bytes()`/`write_bytes()` with `bytes` literals.)
- **Write a new draft page** — heredoc to the remote file:
```bash
ssh llm-wiki 'cat > /srv/workspaces/llm-wiki/wiki/<path>.md' <<'EOF'
---
<canonical v1.1 frontmatter>
---
<body>
EOF
```
- **git on the repo** runs on the CT too: `ssh llm-wiki 'git -C /srv/workspaces/llm-wiki <subcommand>'`.
This depends on `~/.local/bin` being on the LXC's NON-interactive SSH PATH (so `uv`/`uvx` resolve under `ssh host 'cmd'`); that fix lives in the CT's `~/.bashrc` above its interactive guard.
</access_model>
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 299 lines · 82 tokens per session scan A 65fb0592d3c5
up-docs-propagate-wiki is an agent published in the GitHub repository L3DigitalNet/Claude-Code-Plugins (6 stars, last pushed 8d ago), licensed MIT. It adds 82 tokens to every session and 5,156 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
WinForms Expert
Support development of .NET (OOP) WinForms Designer compatible Apps.
Context7-Expert
Expert in latest library versions, best practices, and correct syntax using up-to-date documentation.