Devtu Docs Quality

Documentation Quality Assurance

Two Equal Priorities

Both of the following must be satisfied before an audit is done. Neither overrides the other.

1. Technical correctness — wrong code is worse than no code; it actively breaks users' workflows.

2. Less is more — redundant content is worse than no content; it creates confusion, dilutes trust, and makes maintenance harder.

When in doubt about technical accuracy: verify against source (src/tooluniverse/execute_function.py).

When in doubt about whether content is needed: delete it.

---

Less Is More — The Core Philosophy

ToolUniverse is a serious research tool. Documentation should reflect that with restraint and precision, not volume.

What "less is more" means in practice

Remove, don't add. Every edit should end with fewer words, fewer pages, fewer links. The field moves fast — outdated instructions are actively harmful. When two approaches both work, keep one.

Each concept appears exactly once. If setup instructions appear in three places, a user reading one doesn't know the others exist, and maintainers must update all three on every change. Pick one canonical location and cross-link everywhere else.

No emojis. Emojis in headings, nav items, card titles, and bullet points signal "AI-generated" and undermine trust in a serious scientific tool. Remove them unconditionally — no exceptions for "emphasis."

Shrink, don't summarize. A page reduced from 1000 lines to 30 pointer lines is better than a 1000-line page summarized with a TL;DR box. Do the surgery.

Size budgets (hard limits):

Less-is-more decision tree

When reviewing any piece of content, ask:

1. Does this content exist elsewhere on the site? → Remove it; add a cross-link to the canonical location. 2. Is this content outdated or hard to keep current? → Remove it; link to the upstream source (e.g., aiscientist.tools/setup.md). 3. Can a user reach this information in 2 clicks from the homepage? → If yes and the info is already there, remove this copy. 4. Is this an emoji, decorative header, or filler sentence? → Remove it unconditionally. 5. Is this a "Explore More" card pointing somewhere already in the sidebar? → Remove it.

---

Apple-Style Simplification Rules

Apply these rules to every file touched during an audit.

No emojis in headings, bullet items, card titles, toctree captions, nav bar entries, or button text.

Single setup entry point. The canonical installation path is:

Read https://aiscientist.tools/setup.md and set up ToolUniverse for me.

Per-platform pages must contain exactly three things: (1) official install link, (2) official MCP setup guide link, (3) the setup prompt above. Nothing else — no JSON snippets, no step-by-step instructions, no troubleshooting. Those belong in the setup skill at aiscientist.tools/setup.md.

Consistent tool count. Always "1000+ tools". Never "600+", "750+", "1200+", "10000+", or any other number.

Clean navigation bar. Must not contain "API", "API Keys", or "Contribution" items. Must include a link to https://aiscientist.tools as the first item.

Homepage "Explore More" grid. Maximum 4 cards. Remove any card whose destination is already reachable from the sidebar or the "Get Started" section.

No broken `:doc:` references. Every :doc:some/path` must resolve to an .rst or .md` file on disk. Broken links are never acceptable — remove or replace immediately.

---

Five-Phase Strategy

Run phases in order — D first (instant), then E (simplification scan), then C/A/B.

---

Phase D: Static Method Scan

python3 - <<'EOF'
import re, sys
from pathlib import Path
from collections import defaultdict
DOCS = Path("docs")
EXCLUDE = {"locale", "old", "_build", "__pycache__", "tools", "archive"}
KNOWN_BAD = {
    "list_tools", "run_batch", "run_async", "execute_tool", "call_tool",
    "list_tools_by_category", "configure_api_keys", "get_tool", "get_exposed_name",
    "list_available_methods", "register_tool_from_config", "register_tool",
}
STATIC = [
    (r"\.load_tools\([^)]*(?:use_cache|cache_dir)\s*=", "load_tools() invalid kwargs"),
    (r"ToolUniverse\([^)]*timeout\s*=", "ToolUniverse(timeout=) invalid"),
    (r'"name":\s*"opentarget_', 'old lowercase "opentarget_*" tool name'),
    (r"tu\.run_batch\(", "tu.run_batch() — use tu.run(list, max_workers=N)"),
    (r'\btu\.[A-Z]\w+\s*\(', "tu.ToolName() shorthand — use tu.run({name:...})"),
]
M = re.compile(r'\b(?:tu|tooluni)\.([\w]+)\s*\(')
issues = defaultdict(list)
for f in sorted(list(DOCS.rglob("*.rst")) + list(DOCS.rglob("*.md"))):
    if any(p in f.parts for p in EXCLUDE): continue
    t = f.read_text(errors="replace")
    code = "\n".join(
        re.findall(r"\.\. code-block:: python\n((?:[ \t]+[^\n]*\n|[ \t]*\n)*)", t, re.MULTILINE) +
        re.findall(r"```python\n(.*?)```", t, re.DOTALL))
    if not code.strip(): continue
    rel = str(f.relative_to(DOCS))
    for m in M.finditer(code):
        if not m.group(1).startswith("_") and m.group(1) in KNOWN_BAD:
            issues[rel].append(f"tu.{m.group(1)}()")
    for pat, label in STATIC:
        if re.search(pat, code): issues[rel].append(label)
if issues:
    [print(f"  {f}: {i}") for f in sorted(issues) for i in sorted(set(issues[f]))]
    sys.exit(1)
else:
    print("Phase D clean")
EOF

Fix-or-remove rule — no exceptions:

• Correct replacement exists → fix immediately
• Feature is automatic/internal → remove the call, add prose comment if needed
• Feature doesn't exist → delete the code block or section entirely

Most common fixes:

# wrong → correct
tu.run_batch(list)              → tu.run(list, max_workers=4)
tu.run_async(query)             → await tu.run(query)
tu.call_tool('X', {...})        → tu.run({"name": "X", "arguments": {...}})
tu.execute_tool('X', {...})     → tu.run({"name": "X", "arguments": {...}})
tu.list_tools()                 → tu.list_built_in_tools(mode='list_name')
tu.get_tool('X')                → tu.get_tool_by_name('X')
tu.register_tool(instance)      → tu.register_custom_tool(tool_instance=instance)
tu.register_tool_from_config(c) → tu.register_custom_tool(tool_config=c)
tu.configure_api_keys({})       → REMOVE — use env vars
tu.get_exposed_name(name)       → REMOVE — shortening is automatic
ToolUniverse(timeout=30)        → ToolUniverse()  # no timeout kwarg
load_tools(use_cache=True)      → load_tools()  # no use_cache kwarg
tu.ToolName(key=val)            → tu.run({"name": "ToolName", "arguments": {"key": val}})
opentarget_get_*                → OpenTargets_get_*  (capital O and T)

Special case: Change .. code-block:: python to .. code-block:: text when showing intentionally-wrong code as an error example.

---

Phase E: Simplification Scan

python3 - <<'EOF'
import re, sys
from pathlib import Path

DOCS = Path("docs")
EXCLUDE = {"_build", "__pycache__", "locale"}
issues = []

EMOJI_RE = re.compile(
    r'[\U0001F300-\U0001F9FF\U00002600-\U000027BF\U0001FA00-\U0001FA9F'
    r'\U0001F004\U0001F0CF\U00002702-\U000027B0]'
)

def in_code_block(text, pos):
    """Return True if pos is inside a code block."""
    before = text[:pos]
    # RST code blocks
    rst_starts = [m.end() for m in re.finditer(r'\.\. code-block::[^\n]*\n', before)]
    for start in reversed(rst_starts):
        block = text[start:]
        end_match = re.search(r'\n(?![ \t]|\n)', block)
        block_end = start + (end_match.start() if end_match else len(block))
        if start <= pos <= block_end:
            return True
    # Markdown fenced blocks
    fences = list(re.finditer(r'```', before))
    return len(fences) % 2 == 1

for f in sorted(DOCS.rglob("*.rst")) + sorted(DOCS.rglob("*.md")):
    if any(p in f.parts for p in EXCLUDE):
        continue
    text = f.read_text(errors="replace")
    rel = str(f.relative_to(DOCS))

    # Check emojis in prose
    for m in EMOJI_RE.finditer(text):
        if not in_code_block(text, m.start()):
            line = text[:m.start()].count('\n') + 1
            issues.append(f"EMOJI  {rel}:{line}  {repr(m.group())}")

    # Check wrong tool counts
    for bad in ["600+", "750+", "1200+", "10000+"]:
        if bad in text and "tools" in text[text.find(bad):text.find(bad)+20]:
            issues.append(f"COUNT  {rel}  contains '{bad} tools' (use '1000+')")

    # Check broken :doc: references
    for m in re.finditer(r':doc:`([^`]+)`', text):
        ref = m.group(1).lstrip('/')
        # Strip any display text (format: display <path>)
        if ' <' in ref:
            ref = re.search(r'<([^>]+)>', ref).group(1)
        base = DOCS / Path(ref.replace('/', '/'))
        candidates = [base.with_suffix('.rst'), base.with_suffix('.md'),
                      base / 'index.rst', base / 'index.md']
        if not any(c.exists() for c in candidates):
            line = text[:m.start()].count('\n') + 1
            issues.append(f"LINK   {rel}:{line}  broken :doc:`{ref}`")

# Check navbar in conf.py
conf = Path("docs/conf.py").read_text()
for bad_nav in ['"API"', '"API Keys"', '"Contribution"']:
    if bad_nav in conf:
        issues.append(f"NAV    conf.py  navbar contains {bad_nav}")
if "aiscientist.tools" not in conf:
    issues.append("NAV    conf.py  navbar missing aiscientist.tools link")

# Check per-platform page line counts
platform_dir = DOCS / "guide" / "building_ai_scientists"
for rst in platform_dir.glob("*.rst"):
    if rst.name in ("index.rst", "mcp_support.rst", "compact_mode.rst",
                    "mcpb_introduction.rst", "mcp_name_shortening.rst"):
        continue
    lines = rst.read_text().splitlines()
    if len(lines) > 15:
        issues.append(f"SIZE   guide/building_ai_scientists/{rst.name}  {len(lines)} lines (max 15)")

if issues:
    for i in sorted(issues):
        print(i)
    sys.exit(1)
else:
    print("Phase E clean")
EOF

When Phase E reports issues, apply the less-is-more decision tree above. Fix all items before proceeding to Phase C.

---

Phase C: Live Code Execution

python scripts/test_doc_code_blocks.py

The runner injects a real ToolUniverse instance as preamble, skips blocks needing API keys or async, and classifies NameError on out-of-scope variables as "context-dependent" (not a failure).

Common runtime failures:

---

Phase A: Automated Validation

python scripts/validate_documentation.py

# scripts/validate_documentation.py checks:
DEPRECATED_PATTERNS = [
    (r"python -m tooluniverse\.server", "tooluniverse-server"),
    (r"600\+?\s+tools", "1000+ tools"),
    (r"750\+?\s+tools", "1000+ tools"),
]

---

Phase B: ToolUniverse-Specific Audit

Circular navigation — trace index.rst → quickstart → getting_started manually; no loops allowed.

Tool count — "1000+ tools" consistently everywhere. Run: rg "[0-9]+\+?\s+(tools|integrations)" docs/ --no-filename | sort -u

Auto-generated headers — docs/tools/*_tools.rst and docs/api/*.rst must start with .. AUTO-GENERATED.

CLI docs — every entry under [project.scripts] in pyproject.toml must appear in docs/reference/cli_tools.rst.

Env vars — every os.getenv("TOOLUNIVERSE_*") in src/ must appear in docs/reference/environment_variables.rst.

---

RST Code Block Extractor (important)

Always use re.MULTILINE (not re.DOTALL) for RST blocks to avoid merging adjacent blocks:

# correct
re.findall(r"\.\. code-block:: python\n((?:[ \t]+[^\n]*\n|[ \t]*\n)*)", text, re.MULTILINE)

# wrong — merges adjacent blocks
re.findall(r"\.\..*?code-block.*?python\n((?:[ \t]+.*\n|\n)*)", text, re.DOTALL)

---

Validation Checklist

All items must pass before the audit is done. A partial pass is not acceptable.

Technical correctness:

• [ ] Phase D scan exits 0 — no invalid method calls
• [ ] python scripts/test_doc_code_blocks.py exits 0 — no runtime failures
• [ ] No spec['parameters'] without format="openai"
• [ ] Automated validation passes (0 HIGH issues)
• [ ] All CLIs from pyproject.toml documented
• [ ] No circular navigation

Less is more (Apple-style):

• [ ] Phase E scan exits 0 — no emojis, no wrong counts, no broken links, no oversized pages
• [ ] "1000+ tools" used consistently everywhere (not 600+, 750+, 10000+, etc.)
• [ ] Each per-platform setup page: install link + MCP guide link + setup prompt only (≤15 lines)
• [ ] No duplicate setup instructions anywhere on the site
• [ ] Navbar: no "API", "API Keys", "Contribution"; has aiscientist.tools link
• [ ] Homepage "Explore More" grid: ≤4 cards
• [ ] No concept or section appears verbatim in more than one place
• [ ] No "See Also" / "Related" section with more than 4 links

If any item is failing: stop, fix it, re-run the relevant phase, confirm it passes. Do not proceed to the next item until the current one is clean.

---

Reference Files

• API_REFERENCE.md — valid method signatures, wrong-method table, correct patterns
• DOCS_STRUCTURE.md — per-file audit status for all doc files
• scripts/test_doc_code_blocks.py — Phase C live runner

ToolUniverse API Reference for Doc Auditing

Valid Public Methods (ground truth)

Derive the live list any time with:

import sys; sys.path.insert(0, "src")
from tooluniverse import ToolUniverse
print("\n".join(m for m in sorted(dir(ToolUniverse)) if not m.startswith("_")))

Key methods docs most often reference:

ToolUniverse.__init__ kwargs: tool_files, keep_default_tools, log_level, hooks_enabled, hook_config, hook_type, enable_name_shortening

---

Known-Wrong Methods Table

Every entry below appeared in ToolUniverse docs and does not exist. Fix or remove on sight.

---

Fix-vs-Remove Decision

Wrong information must be fixed or removed. It must never be left as-is.

Wrong call found
    │
    ▼
Correct equivalent method exists?
    ├── YES → Fix to the correct method immediately
    └── NO  → Does the feature conceptually exist?
                  ├── YES (auto/internal) → Remove the call; add a prose comment if needed
                  └── NO  → Delete the code block or section entirely

---

Correct Patterns

# ✅ tool_specification — use format="openai" to get 'parameters' key
spec = tu.tool_specification("Tool_Name", format="openai")
for param, info in spec['parameters']['properties'].items():
    print(param, info['type'])

# ✅ batch execution
results = tu.run([
    {"name": "Tool1", "arguments": {"id": "1"}},
    {"name": "Tool1", "arguments": {"id": "2"}},
], max_workers=4)

# ✅ check if tool loaded
if "My_Tool" in tu.all_tool_dict: ...

# ✅ list tool names
names = tu.list_built_in_tools(mode='list_name')

# ✅ filter by category
tools = tu.filter_tools(include_tool_types=["uniprot"])

# ✅ async context — run() is context-aware
result = await tu.run(query)  # no run_async() needed

# ✅ register custom tool from config dict
tu.register_custom_tool(tool_config=my_config_dict)

---

Context-Aware spec['parameters'] Check

Only flag spec['parameters'] as wrong if format="openai" is not set in the same call:

grep -n -B5 "spec\['parameters'\]" <file>
# Verify tool_specification(..., format="openai") appears above it

ToolUniverse Docs File Structure

122 files total. Legend: ✅ audited clean · 📝 has code blocks · 📄 prose only.

docs/guide/ — User tutorials

docs/guide/building_ai_scientists/ — Platform integrations

docs/guide/hooks/ — Output hooks

docs/expand_tooluniverse/ — Developer guides

docs/help/ — Support

docs/dev_docs/ — Internal developer docs

docs/reference/, docs/about/, docs/api/

Auto-generated (never edit manually)

• docs/tools/*_tools.rst — generated by docs/generate_config_index.py
• docs/api/*.rst — generated by sphinx-apidoc
• docs/locale/ — translation files

Documentation Quality Examples

Example 1: Command Deprecation

Before:

# Installation
python -m tooluniverse.server --config config.json

Detected by validation:

docs/installation.md:45 [HIGH]
  Found: python -m tooluniverse.server
  Should be: tooluniverse-server

After:

# Installation
tooluniverse-server --config config.json

Example 2: Tool Count Standardization

Before: Documentation showed:

• "600+ tools" (intro)
• "750 scientific resources" (features)
• "1195 integrations" (about)

Detected by validation:

docs/index.rst:12 [HIGH]
  Found: 600+ tools
  Should be: 1000+ tools

docs/features.rst:34 [HIGH]
  Found: 750 scientific resources
  Should be: 1000+ tools

After: All changed to "1000+ tools"

Example 3: Circular Navigation

Before:

# docs/quickstart.rst
For complete setup, see :doc:`getting_started`

# docs/getting_started.rst
Prerequisites: Complete :doc:`quickstart` first

Issue: Users can't progress - both pages require the other

After:

# docs/index.rst
**New users start here:** :doc:`quickstart` (5 min) → :doc:`getting_started` (30 min)

# docs/quickstart.rst
**Next:** Continue to :doc:`getting_started` for detailed setup

# docs/getting_started.rst
**Prerequisites:** Basic Python knowledge

Example 4: Auto-Generated Header

Before:

ChEMBL Tools
============

.. automodule:: tooluniverse.chembl_tool

Issue: Users might edit this file directly

After:

.. AUTO-GENERATED - DO NOT EDIT MANUALLY
.. Generated by: docs/generate_config_index.py
.. Last updated: 2024-02-05
.. 
.. To modify, edit source files in src/tooluniverse/data/ and regenerate.

ChEMBL Tools
============

.. automodule:: tooluniverse.chembl_tool

Example 5: Missing CLI Documentation

Detected from pyproject.toml:

[project.scripts]
tooluniverse-server = "tooluniverse.smcp_server:main"
tooluniverse-mcp = "tooluniverse.smcp_server:main"
tooluniverse-expert-feedback = "tooluniverse.expert_feedback:main"
generate-mcp-tools = "tooluniverse.scripts.generate_mcp_tools:main"

Found in docs/reference/cli_tools.rst:

• ✅ tooluniverse-server
• ✅ tooluniverse-mcp
• ❌ tooluniverse-expert-feedback (missing)
• ❌ generate-mcp-tools (missing)

Action: Document the missing CLIs

Example 6: Environment Variables

Discovered from code:

cache_dir = os.getenv("TOOLUNIVERSE_CACHE_DIR", "~/.cache/tooluniverse")
log_level = os.getenv("TOOLUNIVERSE_LOG_LEVEL", "INFO")
api_key = os.getenv("OPENAI_API_KEY")

Create documentation:

Environment Variables
=====================

Cache Configuration
-------------------

TOOLUNIVERSE_CACHE_DIR
  Cache directory location
  **Default:** ``~/.cache/tooluniverse``
  **Example:** ``export TOOLUNIVERSE_CACHE_DIR=/tmp/cache``

Logging
-------

TOOLUNIVERSE_LOG_LEVEL
  Logging verbosity
  **Default:** ``INFO``
  **Options:** ``DEBUG``, ``INFO``, ``WARNING``, ``ERROR``
  **Example:** ``export TOOLUNIVERSE_LOG_LEVEL=DEBUG``

API Keys
--------

OPENAI_API_KEY
  OpenAI API key for LLM features
  **Required:** Yes (for LLM features)
  **Example:** ``export OPENAI_API_KEY=sk-...``

Example 7: Terminology Inconsistency

Before:

# Page 1
Use the API endpoint to query data.

# Page 2
Send requests to the URL.

# Page 3
The route returns JSON data.

# Page 4
Access the path /api/v1/tools.

Detected: 4 terms for same concept

After: Standardized to "API endpoint"

# All pages
Use the API endpoint to query data.
The API endpoint returns JSON data.
Access the API endpoint at /api/v1/tools.

Example 8: Full Validation Run

Command:

python scripts/validate_documentation.py

Output:

Scanning documentation...

❌ Found 15 issues

docs/installation.md:45 [HIGH]
  Found: python -m tooluniverse.server
  Should be: tooluniverse-server

docs/index.rst:12 [HIGH]
  Found: 600+ tools
  Should be: 1000+ tools

docs/features.rst:34 [HIGH]
  Found: 750 tools
  Should be: 1000+ tools

docs/quickstart.rst:23 [MEDIUM]
  Found: API URL
  Should be: API endpoint

... (11 more issues)

Summary:
- HIGH: 8 issues (must fix)
- MEDIUM: 7 issues (should fix)

Run fixes and re-validate.

Example 9: CI/CD Integration Success

GitHub Actions output:

✓ Install dependencies
✓ Run validation
  → python scripts/validate_documentation.py
  → ✅ Documentation validation passed
✓ Check auto-generated headers
  → All files have required headers
✓ Build documentation
  → Sphinx build successful

Example 10: Audit Report

# Documentation Quality Report

**Date**: 2024-02-05
**Scope**: Full ToolUniverse documentation

## Executive Summary
- Files scanned: 127
- Issues found: 23 (Critical: 0, High: 8, Medium: 12, Low: 3)
- Estimated fix time: 3 hours

## High Priority Issues (8)

### 1. Deprecated command references
- **Files affected**: 5
- **Pattern**: `python -m tooluniverse.server`
- **Fix**: Replace with `tooluniverse-server`
- **Effort**: 15 minutes

### 2. Inconsistent tool counts
- **Files affected**: 7
- **Variations**: 600+, 750+, 1195
- **Fix**: Standardize to "1000+"
- **Effort**: 20 minutes

### 3. Missing CLI documentation
- **Undocumented**: tooluniverse-expert-feedback, generate-mcp-tools
- **Fix**: Add to docs/reference/cli_tools.rst
- **Effort**: 45 minutes

## Medium Priority Issues (12)

### 4. Circular navigation
- **Location**: quickstart.rst ↔ getting_started.rst
- **Fix**: Remove back-reference from getting_started
- **Effort**: 10 minutes

### 5. Auto-generated headers missing
- **Files**: 8 in docs/tools/
- **Fix**: Add headers via regeneration script
- **Effort**: 5 minutes

## Recommendations

**Today:**
1. Run validation script and fix all HIGH issues (1.5 hrs)
2. Fix circular navigation (10 min)

**This week:**
3. Document missing CLIs (45 min)
4. Add auto-generated headers (5 min)
5. Create environment variables reference (30 min)

**Next sprint:**
6. Create comprehensive glossary
7. Add validation to CI/CD

## Validation
Run `python scripts/validate_documentation.py` after fixes

Documentation Quality Skill - Merge Summary

What Changed

Merged two complementary documentation skills into one comprehensive skill:

Before (2 separate skills):

1. `devtu-audit-docs` (431 lines)

• ToolUniverse-specific manual audits
• Circular navigation checks
• MCP configuration duplication
• Tool count consistency
• Auto-generated file conflicts

2. `devtu-optimize-docs` (770 lines)

• General-purpose automated validation
• Python validation scripts
• Command accuracy testing
• Terminology consistency
• Generic patterns for any project

After (1 merged skill):

`devtu-docs-quality` (471 lines + examples)

• Phase A: Automated validation (generic, reusable)
• Phase B: ToolUniverse-specific audits
• Combined strengths of both approaches
• Clearer workflow: automation first, then specific checks

Benefits of Merging

1. Single entry point - No confusion about which skill to use 2. Logical workflow - Automated checks first, then specific audits 3. Complete coverage - Both generic and ToolUniverse-specific issues 4. Better organized - Clear phases instead of separate files 5. No duplication - Shared concepts unified

Key Features

Automated Validation (Phase A)

• Python validation script template
• Command accuracy checking
• Link integrity validation
• Terminology consistency tracking
• Smart context-aware pattern matching

ToolUniverse Audits (Phase B)

• Circular navigation detection
• Duplicate content checks
• Tool count standardization ("1000+ tools")
• Auto-generated file headers
• CLI documentation verification
• Environment variables documentation
• Technical jargon glossary checks
• CI/CD regeneration validation

File Structure

devtu-docs-quality/
├── SKILL.md          # Main skill (471 lines)
├── EXAMPLES.md       # 10 concrete examples
└── SUMMARY.md        # This file

Usage Pattern

# Step 1: Run automated validation
python scripts/validate_documentation.py

# Step 2: Fix HIGH priority issues

# Step 3: Run ToolUniverse-specific checks
# - Check circular navigation manually
# - Verify tool count consistency
# - Check auto-generated headers
# - Verify CLI documentation

# Step 4: Re-validate
python scripts/validate_documentation.py

When to Use

Use devtu-docs-quality when:

• ✅ Reviewing documentation before release
• ✅ After major refactoring (commands, APIs, tool counts changed)
• ✅ Users report confusing or outdated documentation
• ✅ Want to establish automated validation pipeline
• ✅ Need to check for circular navigation or structural problems
• ✅ Any documentation quality concern (audit, optimize, fix, review)

Trigger Terms

The skill activates on:

• "audit docs" / "audit documentation"
• "optimize docs" / "optimize documentation"
• "review docs" / "review documentation"
• "fix docs" / "fix documentation"
• "check docs quality"
• "validate documentation"
• "documentation cleanup"

Migration for Users

If you were using either of the old skills:

• All functionality is preserved in devtu-docs-quality
• Workflow is now clearer: Phase A (automated) → Phase B (specific)
• Validation script template included in SKILL.md
• More examples in EXAMPLES.md

Files Removed

• .cursor/skills/devtu-audit-docs/ (merged into devtu-docs-quality)
• .cursor/skills/devtu-optimize-docs/ (merged into devtu-docs-quality)
• skills/devtu-audit-docs/ (merged into devtu-docs-quality)
• skills/devtu-optimize-docs/ (merged into devtu-docs-quality)

Files Created

• .cursor/skills/devtu-docs-quality/SKILL.md
• .cursor/skills/devtu-docs-quality/EXAMPLES.md
• .cursor/skills/devtu-docs-quality/SUMMARY.md
• skills/devtu-docs-quality/ (synchronized copy)