Example Documentation Skill

Goal: create a skill that helps an agent answer and author code for a library without repeatedly re-reading upstream docs.

Input collection approach

This case used breadth-first source collection and only stopped when new retrieval yielded mostly duplicates:

1. Official docs landing pages and navigation trees.
2. All API/class/module reference pages.
3. Configuration and environment reference pages.
4. Official examples/tutorials.
5. Troubleshooting/error catalog pages.
6. Migration/deprecation/changelog pages.
7. Upstream repo README plus canonical examples.
8. In-repo usage of the library (rg on imports and key APIs).

Coverage matrix used

Required dimensions tracked during synthesis:

1. Setup and installation.
2. Core primitives and API surface.
3. Configuration and runtime options.
4. Normal usage patterns.
5. Edge cases and failure handling.
6. Version-specific differences.
7. Migration and deprecation guidance.
8. Instructional templates/examples for direct reuse.

Synthesized artifacts produced

The resulting skill references included:

1. Happy-path implementation template.
2. Production-safe variant with defensive defaults.
3. Anti-pattern and corrected implementation.
4. Intent-to-reference routing guide (which section to load for which user request).
5. Gap log with explicit next retrieval steps.

Source-to-decision trace (sample)

1. Source class: migration/changelog docs. Decision: add a version-compatibility checklist section to the skill. Why: multiple API signatures existed across versions; without this, answers were inconsistent.
2. Source class: troubleshooting/error catalog. Decision: add an error-to-fix lookup table in references. Why: user prompts often start from failures, not idealized setup.
3. Source class: in-repo usage scan (rg). Decision: prioritize examples matching local project patterns. Why: produced outputs became directly usable with fewer edits.

Concrete artifacts (sample)

1. Prompt and output skeleton: Prompt: “Configure client for retries and auth in production.” Output: a production-safe template with retry/backoff, timeout defaults, and auth placeholders.
2. Anti-pattern transformation: Before: single inline config with no timeout/error handling. After: structured config with explicit timeout, retry policy, and failure handling notes.
3. Reference routing snippet: If request mentions “migration” -> load migration/changelog reference first, then API reference.

What made this high quality

1. Input retrieval was exhaustive across all doc classes, not just top pages.
2. The skill shipped transformed examples, not citation-only notes.
3. Coverage and gaps were explicit, so iteration could continue safely.