Replace Module 10's literal /path/to/ placeholders and gloss the unexplained git apply #53

New Issue

2026-06-22T14:24:04-04:00

claude commented

2026-06-22 14:24:04 -04:00

Problem

Unlike the Module 1 exemplar's concrete copy instruction, Module 10 embeds literal cp /path/to/modules/10-…/lab/tasks-app/*.py . and git apply /path/to/…/ai-change.patch, which a literal-minded learner can type verbatim. git apply is a brand-new command (it appears only in M10 across all modules) and is used with no gloss; it silently rewrites two files. Part A also creates a separate review-lab repo distinct from tasks-app, which can confuse.

Evidence

modules/10-reviewing-code-you-didnt-write/README.md Part B step 3 (~line 242): "git apply /path/to/modules/10-reviewing-code-you-didnt-write/lab/ai-change.patch" with no explanation; cp /path/to/... (~line 209). git apply appears nowhere else in the course.

Why it matters

Clarity/consistency gap vs the locked Module 1 exemplar, plus a first-use command with no gloss in a module otherwise careful about its git mechanics.

Proposed change

Replace /path/to/ with a concrete repo-relative instruction in the Module 1 style (or the M5 /path/to/... placeholder convention used elsewhere).
Add a half-sentence gloss on git apply, e.g. "applies the AI's proposed change onto this branch as if it were its PR, so you can review it before merging." (Avoid "simulating its PR landing" — the lab ends in "request changes," not a merge.)
Note that review-lab is a throwaway repo separate from tasks-app.

Acceptance criteria

No paste-ready command contains an unresolved /path/to/ placeholder without explanation.
git apply is glossed on first use.
review-lab is identified as a throwaway repo distinct from tasks-app.

Affected files

modules/10-reviewing-code-you-didnt-write/README.md

References

Source finding F59 (realVotes 2/3 — one lens judged /path/to/ an acceptable course-wide convention for this audience; lower confidence).

Filed from an adversarial multi-agent course review (217 raw findings → 54 adversarially-verified survivors). Scoped for manual review; intentionally not auto-assigned to an agent.

## Problem Unlike the Module 1 exemplar's concrete copy instruction, Module 10 embeds literal `cp /path/to/modules/10-…/lab/tasks-app/*.py .` and `git apply /path/to/…/ai-change.patch`, which a literal-minded learner can type verbatim. `git apply` is a brand-new command (it appears only in M10 across all modules) and is used with no gloss; it silently rewrites two files. Part A also creates a separate review-lab repo distinct from `tasks-app`, which can confuse. ## Evidence `modules/10-reviewing-code-you-didnt-write/README.md` Part B step 3 (~line 242): "git apply /path/to/modules/10-reviewing-code-you-didnt-write/lab/ai-change.patch" with no explanation; `cp /path/to/...` (~line 209). `git apply` appears nowhere else in the course. ## Why it matters Clarity/consistency gap vs the locked Module 1 exemplar, plus a first-use command with no gloss in a module otherwise careful about its git mechanics. ## Proposed change 1. Replace `/path/to/` with a concrete repo-relative instruction in the Module 1 style (or the M5 `/path/to/...` placeholder convention used elsewhere). 2. Add a half-sentence gloss on `git apply`, e.g. "applies the AI's proposed change onto this branch as if it were its PR, so you can review it before merging." (Avoid "simulating its PR landing" — the lab ends in "request changes," not a merge.) 3. Note that review-lab is a throwaway repo separate from `tasks-app`. ## Acceptance criteria - [ ] No paste-ready command contains an unresolved `/path/to/` placeholder without explanation. - [ ] `git apply` is glossed on first use. - [ ] review-lab is identified as a throwaway repo distinct from tasks-app. ## Affected files - `modules/10-reviewing-code-you-didnt-write/README.md` ## References Source finding F59 (realVotes 2/3 — one lens judged `/path/to/` an acceptable course-wide convention for this audience; lower confidence). --- *Filed from an adversarial multi-agent course review (217 raw findings → 54 adversarially-verified survivors). Scoped for manual review; intentionally not auto-assigned to an agent.*