Atlas / Skills / agenticnotetaking / Validate

ValidateSAFE

skills/agenticnotetaking/validate

Claude Code plugin that generates individualized knowledge systems from conversation. You describe how you think and work, have a conversation and get a complete second brain as markdown files you own.

Verdict
SAFE
Grade
B
Trust score
89 /100
Version
—
Hosts
—
License
MIT
Stars
3,492
01

Overview

Claude Code plugin that generates individualized knowledge systems from conversation. You describe how you think and work, have a conversation and get a complete second brain as markdown files you own.

Read from source at commit 9eccf964e924OBSERVED · 2026-10-07
02

What it tells the agent

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

---
name: validate
description: Schema validation for notes. Checks against domain-specific templates. Validates required fields, enum values, description quality, and link health. Non-blocking — warns but doesn't prevent capture. Triggers on "/validate", "/validate [note]", "check schema", "validate note", "validate all".
user-invocable: true
allowed-tools: Read, Grep, Glob
context: fork
model: sonnet
---

## Runtime Configuration (Step 0 — before any processing)

Read these files to configure domain-specific behavior:

1. **`ops/derivation-manifest.md`** — vocabulary mapping, platform hints
   - Use `vocabulary.notes` for the notes folder name
   - Use `vocabulary.note` / `vocabulary.note_plural` for note type references
   - Use `vocabulary.topic_map` for MOC references
   - Use `vocabulary.templates` for the templates folder path

2. **`ops/config.yaml`** — processing depth
   - `processing.depth`: deep | standard | quick

If these files don't exist, use universal defaults.

**Processing depth adaptation:**

| Depth | Validation Behavior |
|-------|---------------------|
| deep | Full schema validation. All checks enabled including composability analysis and cross-reference verification |
| standard | Full validation — all checks enabled |
| quick | Basic schema validation only — required fields, YAML validity, enum values |

## EXECUTE NOW

**Target: $ARGUMENTS**

Parse immediately:
- If target contains a note name: validate that specific note
- If target contains `--handoff`: output RALPH HANDOFF block at end
- If target is "all" or "notes": validate all notes in {DOMAIN:notes}/ directory
- If target is empty: ask which note to validate

**Execute these steps:**

### Step 1: Locate Template

Determine which template applies to the target note:

1. Check the note's location — notes in {DOMAIN:notes}/ use the standard note template
2. Check the `type` field in frontmatter — specialized types may have dedicated templates
3. Look for a templates directory (check `ops/templates/` or domain-specific path from derivation manifest)
4. If the template has a `_schema` block, read it — this is the authoritative schema definition

If no template is found, use the default schema checks below.

### Step 2: Read Target Note

Read the target note's full YAML frontmatter. Parse:
- All YAML fields and their values
- The body content (for link scanning)
- The footer section (for Topics and Relevant Notes)

### Step 3: Run Schema Checks

Run ALL validation checks. Each check produces PASS, WARN, or FAIL.

**START NOW.**

---

## Schema Checks

### Required Fields (FAIL if missing)

| Check | Rule | How to Verify |
|-------|------|---------------|
| `description` | Must exist and be non-empty | Check YAML frontmatter for `description:` field with non-empty value |
| Topics | Must link to at least one {DOMAIN:topic map} | Check for `topics:` in YAML or `Topics:` section in footer. Must contain at least one wiki link |

A missing required field is a hard failure. The note cannot pass validation without these.

### Description Quality (WARN if weak)

| Check | Rule | How to Verify |
|-------|------|---------------|
| Length | Should be ~50-200 characters | Count characters in description value |
| New information | Must add context beyond the title | Compare description text against filename/title — if semantically equivalent, WARN |
| No trailing period | Convention: descriptions don't end with periods | Check last character |
| Single sentence | Should be one coherent statement | Check for sentence-ending punctuation mid-description |

**How to check "adds new info":** Read the title (filename without .md). Read the description. If the description merely restates the title using different words, it fails this check. A good description adds one of:
- **Mechanism** — how or why the claim works
- **Scope** — what boundaries the claim has
- **Implication** — what follows from the claim
- **Context** — where the claim applies

**Examples:**

Bad (restates title):
- Title: `vector proximity measures surface overlap not deep connection`
- Description: "Semantic similarity captures surface-level overlap rather than genuine conceptual relationships"

Good (adds mechanism):
- Title: `vector proximity measures surface overlap not deep connection`
- Description: "Two notes about the same concept with different vocabulary score high, while genuinely related ideas across domains score low"

### YAML Validity (FAIL if broken)

| Check | Rule | How to Verify |
|-------|------|---------------|
| Frontmatter delimiters | Must start with `---` on line 1 and close with `---` | Read first line and scan for closing delimiter |
| Valid YAML | Must parse without errors | Check for common YAML errors: unquoted colons in values, mismatched quotes, bad indentation |
| No duplicate keys | Each YAML key appears only once | Scan for repeated field names |
| No unknown fields | Fields not in the template schema | Compare against `_schema.required` and `_schema.optional` if available — unknown fields get WARN |

### Domain-Specific Enum Checks (WARN if invalid)

If the note has fields with enumerated values, check them against the template's `_schema.enums` block:

| Field | Expected | Severity |
|-------|----------|----------|
| `type` | Values from template enum (e.g., claim, methodology, tension, problem, learning) | WARN |
| `status` | Values from template enum (e.g., preliminary, open, dissolved) | WARN |
| `classification` | Values from template enum (e.g., open, closed) | WARN |
| Custom domain fields | Values from template enum | WARN |

If a field has a value not in the enum list, report the invalid value and list the valid options.

### Link Health (WARN per broken link)

| Check | Rule | How to Verify |
|-------|------|---------------|
| Body wiki-links | Each `[[link]]` should point to an existing file | Extract all `[[...]]` patterns from body, check each against file tree |
| Topics links | {DOMAIN:topic map} referenced 
03

Trust audit

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

LayerWhat it checksResult
L0Provenance & inventoryPASS
L1Static analysis of the codeNA
L2Instruction surface (what it tells the agent)PASS
L3Class-specific surfacePASS
L4Behavioural (sandbox)SKIPPED

What the source does

Filesystem
none-observed
Network
none-observed
Shell
none-observed
Dependencies
pinned
Secrets in source
none-found

Findings (0)

No findings outside the package's declared scope.

Gates applied: no_behavioural_pass.

Audited 2026-10-07 · audit v0.4.1 · source sha 9eccf964e924full audit observations/trust-audit/skill/agenticnotetaking__validate.json · Report an issue / request a re-scan
04

Audit history

Every audit this skill has had.

DateSourceVerdictGradeScoreChange
2026-10-079eccf964e924SAFEB89first audit
05

Questions

What does the Validate skill do?

Claude Code plugin that generates individualized knowledge systems from conversation. You describe how you think and work, have a conversation and get a complete second brain as markdown files you own.

Is Validate safe to install?

The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.

What can Validate access on my machine?

The audit observed no filesystem, network or shell use at all in its source.

How current is this page?

The grade is for one exact copy of the source (9eccf964e924), read on 2026-10-07. The repository is watched, and a new audit runs when it changes — this is the first audit.

Advertisement