Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add saajunaid/caddis-plugin --skill confluence-hlagit clone --depth 1 https://github.com/saajunaid/caddis-pluginWrote 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/skills/saajunaid/caddis-plugin/confluence-hla)<a href="https://agentmods.dev/skills/saajunaid/caddis-plugin/confluence-hla"><img src="https://agentmods.dev/badge/skills/saajunaid/caddis-plugin/confluence-hla/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/saajunaid/caddis-plugin/confluence-hla"><img src="https://agentmods.dev/badge/skills/saajunaid/caddis-plugin/confluence-hla.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00040 | $0.07630 |
| Opus 5.5 | $0.00016 | $0.03052 |
| Sonnet 5.5 | $0.00008 | $0.01526 |
| Haiku 4.5 | $0.00004 | $0.00763 |
Grade A, and why
confluence-hla 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 19d 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 — 382 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Confluence High-Level Architecture (HLA) Authoring Skill
A cross-agent skill for Claude Code, OpenAI Codex CLI, and Antigravity (agy) to research systems, discover data pipelines, author enterprise-grade High-Level Architecture (HLA) documents, and export them directly into clean, self-contained Confluence Storage Format XHTML with native PlantUML macros.
1. Core Principles
- Simplified Technical English (ASD-STE100 Style):
- Short sentences (under 20 words where possible).
- Active voice ("The worker queries telemetry" rather than "Telemetry is queried by the worker").
- Clear, everyday technical words; one concept per sentence.
- CRITICAL: Apply this style naturally. NEVER write notices, disclaimers, or metadata banners into the document stating that it was written in Simplified Technical English or ASD-STE100.
- Zero AI Fingerprints, Zero Em-Dashes & Anti-Slop Discipline:
- NEVER use em-dashes (
—) or en-dashes (–): Em-dashes are the single most prominent dead giveaway of LLM-generated prose. Enterprise solutions architects do not write em-dashes. Use standard colons (:), hyphens (-), parentheses ((...)), or separate sentences instead.- Banned:
High-Level Architecture (HLA) — AppointmentAssist - Required:
High-Level Architecture (HLA): AppointmentAssist - Banned:
- **Cancel Appointment — Area Outage:** ... - Required:
- **Cancel Appointment (Area Outage):** ... - Banned:
The system — which runs on Windows Server — polls telemetry. - Required:
The system runs on Windows Server. It polls telemetry.
- Banned:
- Absolute Ban on AI Slop Buzzwords and Hyperbolic Clichés:
- Banned verbs & nouns:
delve,tapestry,plethora,beacon,game-changer,testament,revolutionize,cornerstone,leverage(useuse,read, orquery),foster,streamline,harnessing,embark,landscape(in non-geographical contexts). - Banned adjectives:
pivotal,crucial,meticulous,seamless,holistic,robust,comprehensive,cutting-edge,paramount,vibrant. - Banned transitions & framing:
Furthermore,Moreover,In conclusion,It is important to note,Let's explore,Here is a breakdown:,Key takeaways:,Summary:.
- Banned verbs & nouns:
- Clean ASCII Punctuation:
- Never use curly / smart quotes (
“,”,‘,’) — always use straight ASCII quotes (",'). - Never use unicode bullet symbols (
•) — always use standard markdown hyphens (-). - Never use backtick formatting on ordinary English nouns.
- Never use curly / smart quotes (
- Zero Meta-Commentary & Self-Referential Announcements:
- Never write phrases announcing what the document is doing (e.g., "In this document we delve into...", "As requested by the user...").
- Start immediately with the title, status, metadata block, and executive overview.
- Zero Tool / Agent / Assistant / Model Self-Attribution:
- The final Confluence export must NEVER contain YAML frontmatter, agent attribution tags, model names, preparation notices, or document references that expose internal repository paths or harness tooling (
caddis,claude code,codex,antigravity,asd-ste100). - Note: Production application components (such as a local LLM daemon running on an on-premises GPU server) are documented as objective production system components, but the HLA itself must never attribute its own authorship to AI.
- The final Confluence export must NEVER contain YAML frontmatter, agent attribution tags, model names, preparation notices, or document references that expose internal repository paths or harness tooling (
- Automated Converter Enforcement:
- The converter script (
convert_hla_to_confluence.py) enforces this automatically by stripping all em-dashes, en-dashes, smart quotes, and AI buzzwords during HTML export.
- The converter script (
- NEVER use em-dashes (
- Confluence Native Compatibility (Self-Contained PlantUML):
- Uses native Confluence Storage Format XHTML.
- All tables wrapped with
<table class="wrapped">and explicit<colgroup>. - Table of contents generated via
<ac:structured-macro ac:name="toc" ac:schema-version="1"/>. - Pure PlantUML: Diagrams are rendered natively via Confluence PlantUML macros. No image uploads, attachments, or base64 data URIs are required—the generated HTML is 100% self-contained and ready for a single copy-paste into the Confluence Source Editor.
- Guaranteed 100% valid XML parseable by the Confluence Source Editor.
- Mermaid & PlantUML Full-Text Fidelity (Zero Truncation Rule):
- In Markdown (
.md), always wrap Mermaid text lines using explicit<br/>tags (maximum ~35 to 45 characters per line) across actors, notes, messages, node labels, and edge transitions. - Never allow unwrapped single-line strings to blow out diagram dimensions (>1200px) or truncate text in Git viewers, VS Code, or Confluence viewports.
- The converter (
convert_hla_to_confluence.py) dynamically maps<br/>to\nin PlantUML so that both formats render complete, untruncated text cleanly.
- In Markdown (
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 19d ago Changed · -27 tokens per session fc6050c78d2b
- 1mo ago First seen · 382 lines · 67 tokens per session scan A 83250c1db785
confluence-hla is a skill published in the GitHub repository saajunaid/caddis-plugin (0 stars, last pushed 2d ago), licensed MIT. It adds 40 tokens to every session and 7,630 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-09-16.
Other skills, from other repositories
docx
Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of "Word doc", "word document", ".docx", or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when…
pptx
Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying…
xlsx
Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or…
Use this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and…
document-processor
Guidance for processing documents, extracting content, and transforming structured information. Use when the user asks to process, parse, extract, or transform document content such as PDFs, Word files, or spreadsheets.
attendance-signin-sheet
A spreadsheet generator for printable attendance or meeting sign-in sheets. It turns a list of names into rows with columns such as signatures, dates, departments, or employee IDs.