fix: enhance using-ui-bundle-salesforce-data skill @W-21686611 [post-tdx]

[channiganfsu]

@W-21686611@
References: Contributing guide · Skill authoring guide · Agent Skills spec

What changed

Improved the using-webapplication-salesforce-data skill's description, body structure, directory layout, and internal path consistency:

• Rewrote the frontmatter description from ~103 words to ~67 words using a "what + when + not" two-sentence pattern
• Added "When This Skill Activates" and "Preconditions" sections to the SKILL.md body
• Renamed references/ to docs/ and updated all internal pointers in SKILL.md
• Fixed stale script paths (using-webapp- → using-webapplication-) in schema-introspection.md and graphql-search.sh
• Added trigger evals (10 should-trigger + 10 should-not-trigger) and 3 end-to-end evals

Why

The description was too long and lacked a negative boundary, making it prone to misfires on adjacent domains (LWC, Flows, Bulk API, Apex triggers, metadata deployment). The body jumped straight into technical detail without a clear activation list or consolidated preconditions table, which hurts post-trigger alignment. The stale paths in the reference file and script would cause runtime failures when following the documented instructions.

Notes

• Description changes were grounded in a research report that analyzed 29 production skills across major ecosystems and found the "what + when + not" pattern to be the highest-signal description style
• The trigger eval set (20 total) is a starting point — the research recommends 40-100 for full coverage
• The docs/ directory at the repo root (research report) is untracked and not included in this PR

---

Skills

Manual checklist

Description quality

• Describes what the skill does and the expected output
• Includes relevant Salesforce domain keywords (Apex, LWC, SOQL, metadata types, etc.)
• Trigger phrases are specific enough for Vibes to select this skill reliably
  Instructions
• Clear goal statement
• Step-by-step workflow
• Validation rules for generated output
• Defined output / artifact
  Context efficiency
• Core instructions are concise — supporting material lives in docs/ subdirectory
• No unnecessary background explanation in the body

Automated checks

Enforced by CI (npm run validate:skills) per the Agent Skills spec:

• Directory is one level deep, named in kebab-case (max 64 chars), contains SKILL.md
• Frontmatter name matches directory name; description is present, ≥ 20 words, ≤ 1024 characters, and includes trigger language
• Body is non-empty and under 500 lines
• Name uses gerund form ⚠ (warning — does not block merge)

[channiganfsu]

These were generated based on the /skill-creator skill feedback. We can leave them in, or remove them. I like the idea, but I haven't thoroughly vetted them yet.

[channiganfsu]

Notice how the script doesn't output Delete related wrappers and fields, because it is essentially a unique case in what is required as input and output