docs: strengthen PET Rust review guidance (Fixes #496)

[Karthik Nadig (karthiknadig)]

Summary

• add a PET-wide Rust coding skill alongside locator-specific guidance
• capture path identity/cache, Unicode-safe parsing, hot-path I/O/allocation, and cross-platform rules
• require tests to prove claimed read-count, cache-hit, and event-count invariants
• wire the recurring checks into the Reviewer agent for every Rust change

Review retrospective

Recent feedback clustered around:

• normalized cache keys and caller-facing path preservation (perf: reuse Conda environment metadata across refreshes (Fixes #476) #487, fix: normalize Conda environment cache keys (Fixes #488) #490)
• Unicode-safe byte indexing and ASCII format handling (perf: reuse Conda history while parsing environments (Fixes #475) #493)
• proving optimization scope and read counts (perf: reuse Conda history while parsing environments (Fixes #475) #493)
• duplicate side effects and precise pattern semantics (feat: improve recursive environment glob diagnostics (Fixes #427) #495)
• workflow assignment/merge postconditions, already addressed in chore: harden maintainer completion workflow (Fixes #485) #486

Fixes #496

[github-actions]

Performance Report (Linux) ➖

Metric	PR (P50)	PR (P95)	Baseline (P50)	Delta	Change
Server Startup	1ms	1ms	1ms	0ms	0%
Full Refresh	67ms	290ms	55ms	12ms	20.0%

Results based on 10 iterations. P50 = median, P95 = 95th percentile.

---

Legend

• 🚀 Significant speedup (>100ms faster)
• ✅ Faster than baseline
• ➖ No significant change
• 🔺 Slower than baseline (>100ms)
• ⚠️ Significant slowdown (>500ms)

[github-actions]

Performance Report (macOS)

Metric	PR (P50)	PR (P95)	Baseline (P50)	Delta
Server Startup	60ms	615ms	86ms	-26ms
Full Refresh	101ms	30240ms	140ms	-39ms

Results based on 10 iterations. P50 = median, P95 = 95th percentile.

---

Legend

• 🚀 Significant speedup (>100ms faster)
• ✅ Faster than baseline
• ➖ No significant change
• 🔺 Slower than baseline (>100ms)
• ⚠️ Significant slowdown (>500ms)

[github-actions]

Test Coverage Report (Linux)

Metric	Value
Current Coverage	80.6%
Base Branch Coverage	80.6%
Delta	0% ➖

---

Coverage unchanged.

[github-actions]

Performance Report (Windows) ✅

Metric	PR (P50)	PR (P95)	Baseline (P50)	Delta	Change
Server Startup	6ms	9ms	9ms	-3ms	-33.3%
Full Refresh	106ms	520ms	133ms	-27ms	-20.3%

Results based on 10 iterations. P50 = median, P95 = 95th percentile.

---

Legend

• 🚀 Significant speedup (>100ms faster)
• ✅ Faster than baseline
• ➖ No significant change
• 🔺 Slower than baseline (>100ms)
• ⚠️ Significant slowdown (>500ms)

[github-actions]

Test Coverage Report (Windows)

Metric	Value
Current Coverage	77.52%
Base Branch Coverage	77.31%
Delta	0.21% ✅

---

Coverage increased! Great work!

[Copilot]

Pull request overview

This PR strengthens PET’s Rust review guidance by adding a new PET-wide Rust coding skill and updating the Reviewer agent checklist so Rust reviews consistently cover recent recurring issues (cache key/path identity, Unicode-safe parsing, hot-path I/O/allocation, and “tests must prove invariants”).

Changes:

• Add a new .github/skills/rust-coding-skill/SKILL.md with PET-wide Rust correctness/performance guidance and test-invariant expectations.
• Update .github/agents/Reviewer.agent.md to apply the new Rust coding skill for any Rust-changing PRs and to expand the Rust review checklist accordingly.

Show a summary per file

File	Description
.github/skills/rust-coding-skill/SKILL.md	Introduces PET-wide Rust guidance (caches/path identity, parsing safety, hot paths, locks, and test invariants).
.github/agents/Reviewer.agent.md	Expands reviewer checklist and ensures Rust reviews load/apply the new Rust coding skill.

Review details

• Files reviewed: 2/2 changed files
• Comments generated: 4
• Review effort level: Lite

[Copilot]

Review details

• Files reviewed: 2/2 changed files
• Comments generated: 0 new
• Review effort level: Lite

[Copilot]

Review details

Suppressed comments (2)

.github/agents/Reviewer.agent.md:173

• This line references the coding guidance by file path, but the skills are invoked by name. Referring to the skill name keeps the instruction stable if the file is moved/renamed and aligns with the surrounding skill documentation.

Apply `.github/skills/rust-coding-skill/SKILL.md` to every Rust review, not only locator changes.

.github/agents/Reviewer.agent.md:51

• The reviewer checklist references skills by file path, which is more brittle and less actionable than using the skill names (the agent invokes skills by name). Using consistent skill-name references also matches how other skill docs refer to each other.

This issue also appears on line 173 of the same file.

- If any changed file is Rust, load and apply both `.github/skills/rust-coding-skill/SKILL.md` and `.github/skills/rust-locator-patterns/SKILL.md`.

• Files reviewed: 2/2 changed files
• Comments generated: 0 new
• Review effort level: Lite

[Copilot]

Review details

Suppressed comments (2)

.github/agents/Reviewer.agent.md:52

• The new guidance says to load both rust-coding-skill and rust-locator-patterns whenever any changed file is Rust, but later this doc only mandates applying rust-coding-skill for all Rust reviews. To keep instructions consistent (and avoid over-applying locator-order rules to non-locator Rust changes), consider always loading rust-coding-skill, and loading rust-locator-patterns only when locator/discovery/identification code is in scope.

Before reading code:

- If any changed file is Rust, load and apply both the `rust-coding-skill` and `rust-locator-patterns` skills.

.github/agents/Reviewer.agent.md:174

• Minor wording issue: "Apply the rust-coding-skill skill" is redundant. Dropping the second "skill" reads more cleanly and avoids repeating the term in this heading text.

### General Rust Correctness and Performance

Apply the `rust-coding-skill` skill to every Rust review, not only locator changes.

• Files reviewed: 2/2 changed files
• Comments generated: 0 new
• Review effort level: Lite

[Copilot]

Review details

• Files reviewed: 2/2 changed files
• Comments generated: 0 new
• Review effort level: Lite