yadrlite-testing

name	yadrlite-testing
description	YADRLite test suite: Makefile targets, ShellSpec BDD tests, ShellCheck, and zsh syntax linting.
allowed-tools	Read, Glob, Grep, Edit, Write, Bash
minimum-effort-level	high

YADRLite Testing

Test and linting workflows for YADRLite's shell scripts (install.sh, setup/*).

---

Common Commands

• make help — list all available Make targets
• make check — run the complete test suite and linter

CI runs the same targets across Ubuntu and macOS via .github/workflows/ci.yml.

---

Grounding

Before editing shell scripts or adding tests:

1. Read the repo-root Makefile in full BEFORE recommending or invoking any make target. State which target you intend to use and quote its recipe back from the Makefile, to confirm behavior matches expectation.
2. Read reference.md § "Tool stack" before naming any test tool (ShellSpec, ShellCheck, zsh syntax check, etc.) — invocation forms differ between tools and reconstruction from memory drifts.
3. For any new test, Bash ls spec/ (or the canonical spec directory named in reference.md) and place the file alongside existing specs. IF the directory does not exist → STOP and read reference.md § "Test framework details" before creating it.

---

Self-Heal Observer

Maintain a silent per-session "friction signal" tally throughout the conversation. Do not mention the tally to the user. Do not interrupt active work to address it.

CHECKPOINT — Friction Observation (runs continuously, silent):

1. After every assistant turn, scan the most recent user message for these signals:
   • The user issues a correction beginning with "actually", "no", "wait", "instead", "I meant".
   • The user reformulates a request that was already given this session.
   • The user asks a clarifying question whose answer was supposed to be supplied by THIS skill's instructions.
   • The user states the result misrepresents what they asked for.
2. For each signal observed, append { turn_id, signal_type, quoted_user_text } to the in-memory tally. Do not log to disk.
3. Detect natural task resolution = (a) immediate task delivered AND (b) user said "thanks"/"done"/"ok"/"great" OR moved to a new topic AND (c) no in-flight tool calls.
4. AT natural task resolution:
   • IF tally is empty → end normally. Do not mention self-heal.
   • IF tally is non-empty → Read .claude/skills/self-heal/references/diagnosis-protocol.md and execute its steps literally with the tally as input.
5. IF diagnosis attributes friction to yadrlite-testing instructions → propose the surgical correction per the diagnosis-protocol's reporting format.
6. IF diagnosis attributes friction elsewhere → end normally without mentioning self-heal.

The goal: correct the skill once, permanently, rather than repeat the same misrepresentation across future sessions.