Contextual guidance and command discovery. Three modes — narrative (first-time), contextual (mid-task), compact (quick reference). Shows available commands, active skills, and intelligent suggestions based on vault state. Triggers on "/help", "what can I do", "show commands",
$ npx -y skills add agenticnotetaking/arscontexta --skill help --agent claude-code
Installs just this skill. Get the whole plugin for auto-invocation.
⚡ How it fires
How this skill gets triggered: by you, by Claude, or both.
Fires itselfClaude auto-loads it when your prompt matches the work.
You can call itInvoke it directly when you want it.
Slash command/help
👁️ Context preview
The summary Claude sees to decide when to auto-load this skill.
Contextual guidance and command discovery. Three modes — narrative (first-time), contextual (mid-task), compact (quick reference). Shows available commands, active skills, and intelligent suggestions based on vault state. Triggers on "/help", "what can I do", "show commands",
📊 Stats
Stars3,451
Forks218
LanguageShell
LicenseMIT
📦 Ships with arscontexta
</> SKILL.md
help.SKILL.md
---name: help
description: Contextual guidance and command discovery. Three modes — narrative (first-time), contextual (mid-task), compact (quick reference). Shows available commands, active skills, and intelligent suggestions based on vault state. Triggers on "/help", "what can I do", "show commands", "how does this work".
user-invocable: true
allowed-tools: Read, Grep, Glob, Bash
version: "1.0"
generated_from: "arscontexta-v1.6"
---## EXECUTE NOW
**Target: $ARGUMENTS**
Parse immediately:
- If target contains `--compact`: force compact mode regardless of vault state
- If target contains a skill name (e.g., "help reduce" or "help reflect"): show detailed help for that specific skill
- If target is empty: determine mode from vault state
**Execute these steps:**
1. Gather vault state (Step 1)
2. Resolve domain vocabulary (Step 2)
3. Determine mode (Step 3)
4. Discover commands dynamically (Step 4)
5. Render the appropriate mode (Step 5)
**START NOW.** Reference below defines each step.
---## Step 1: Gather Vault State
Determine the **notes folder** by checking which domain-named directory exists (notes/, reflections/, concepts/, ideas/, decisions/, memories/, or any custom name from derivation-manifest.md). Then gather counts:
| User hasn't tried /learn | `Grow your graph: /learn [topic related to their domain]` |
| Methodology folder has friction notes | `Your system is learning from friction -- review with /{rethink}` |
| Methodology folder has notes beyond derivation-rationale | `Your system has [N] operational learnings -- query them with /ask or browse ops/methodology/` |
| General orientation needed | [[getting-started]] |
Add: `For details, see manual/{page}.md`
**"Commands you might not know" logic:**
Track which commands the user has likely used by checking for their artifacts:
- /learn used if `ops/sessions/` has learn-related transcripts or inbox has research files
- /remember used if `ops/methodology/` or `ops/observations/` has files
- /graph used if... (no artifact, always suggest if note count > 20)
- /stats used if... (no artifact, always suggest if note count > 10)
Show up to 3 commands the user probably has not tried. Skip commands they clearly have used.
### COMPACT MODE (--compact flag)
Quick reference. No state analysis. No suggestions. Just the commands.
```
--=={ ars contexta : help --compact }==--
/ask [question] Query knowledge system
/learn [topic] Research and grow graph
/next Next recommended action
/{reduce} [source] Extract insights
/{reflect} Find connections
/{reweave} Update older notes
/verify [note] Quality verification
/health Vault diagnostics
/{rethink} Challenge assumptions
/remember Capture methodology learning
/architect Evolution advice
/reseed Restructure system
/add-domain Add new domain
/upgrade Check for improvements
/tutorial Interactive walkthrough
/graph Graph exploration
/stats Vault metrics
/help This guide
```
### SKILL DETAIL MODE (help [skill-name])
When the user asks for help with a specific skill (e.g., `/help reduce` or `/help reflect`):
1. Find the skill's SKILL.md file
2. Read its frontmatter (name, description) and first section
3. Display a focused guide:
```
--=={ ars contexta : help /{skill-name} }==--
{Description from frontmatter}
Usage:
/{skill-name} [arguments]
Examples:
/{skill-name} inbox/article.md
/{skill-name} --since 7d
What it does:
{2-3 sentence summary extracted from the skill's
philosophy or first methodology section}
Related:
/{related-skill} {one-line description}
```
**Special case for /help ask:** Expand the "What it does" section to mention both knowledge layers:
```
What it does:
Queries the bundled methodology knowledge base (249
research claims) AND your local methodology notes
in ops/methodology/. Answers are grounded in
specific claims and applied to your system's
configuration.
```
Extract the pipeline position (what comes before and after this skill) to show workflow context.
**Manual cross-reference:** If `manual/skills.md` exists, also check it for additional context about the requested skill. The manual may contain domain-specific examples and workflow context that the SKILL.md frontmatter does not include.
---
## Rendering Rules
- Max width: 76 characters
- No ANSI color codes
- No emoji
- Monospaced alignment assumed
- Display short command forms (`/reduce`), not plugin-qualified forms (`/arscontexta:reduce`)
- Domain-native names in curly braces (e.g., `/{reduce}`) are resolved from derivation-manifest.md
- When domain name equals universal name, drop the braces
## Edge Cases
**No ops/derivation-manifest.md:** Use universal vocabulary for all command names.
**Pre-init invocation:** If no vault structure exists at all (no notes folder, no ops/), show a minimal message:
```
Ars Contexta is not initialized in this directory.
Run /setup to create your knowledge system.
```
**Multiple domains:** If `ops/derivation-manifest.md` lists multiple domains, show commands grouped by domain in contextual mode.
