# Eval Template

> <!-- Generated by agnostic-ai -->

- Skill: `phel-lang/eval-template` (Agent Skill, multi-file: 3 files)
- Install (CLI): `npx skillmds@latest add phel-lang/eval-template`
- Raw SKILL.md: https://api.skillmd.com/api/skills/phel-lang/eval-template/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: phel-lang (https://skillmd.com/u/phel-lang)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/phel-lang/eval-template

---

<!-- Generated by agnostic-ai -->

# Code to Spec Self Evaluation

After generating the reverse-spec draft, evaluate it against:

## Structure
- [ ] Evidence section present
- [ ] Acceptance Criteria >= 3 (unless truly trivial)
- [ ] Unknowns & Questions present

## Grounding
- [ ] All claims traceable to evidence (file:symbol, config key, or test)
- [ ] No invented PSI types, tokens, grammar rules, extension points, or Phel functions
- [ ] Uncertain areas labeled

## Plugin behavior safety (if applicable)
- [ ] Lexer/parser ordering and malformed-input handling described
- [ ] PSI tree shape / key element types identified
- [ ] Threading and cancellation behavior stated
- [ ] Version-compat and generated-source/registry impact noted

## Observability
- [ ] Logging points defined where failures can occur
- [ ] Existing test coverage identified

If any item remains unchecked, revise the draft before declaring it complete.