Slash Command

/find-wasted-code

Statically scans a repo for dead-end tool calls, broken links, missing helpers, and footgun configs, then writes a tracked report to docs/wasted-code/report.md.

code-quality

Popularity

Stars

Invocation

How this command is triggered — by the user, by Claude, or both

Slash command

/fusebase-flow:find-wasted-code

Model invocable

No pre-commands

Configuration

Namespacecommands/

Context Preview

The summary Claude sees in its command listing — used to decide when to auto-load this command

# /find-wasted-code

Invoke the **find-wasted-code** skill (`flow-skills/find-wasted-code/SKILL.md`).

1. Run `python hooks/local/find-wasted-code.py` (if `python` is missing, run `python3 hooks/local/find-wasted-code.py`). Optional args: `--print` (summary only, no write), `--date YYYY-MM-DD` (deterministic in-report date), `--root PATH`. Self-check: `--selftest` runs the per-rule golden fixtures (adversarial slug corpus, W5 baseline, scope exclusions, output containment, secret-scanner coexistence).
2. Read the report it wrote (`docs/wasted-code/report.md`) — per-rule findings (W1 dead-en...

Command Content

16 lines · ~675 tokens

Stats

LanguageShell

Stars6

MaintenanceExcellent

Last CommitJul 26, 2026

Actions

View Source View Plugin View on GitHub View README

/find-wasted-code

Invoke the find-wasted-code skill (flow-skills/find-wasted-code/SKILL.md).

Run python hooks/local/find-wasted-code.py (if python is missing, run python3 hooks/local/find-wasted-code.py). Optional args: --print (summary only, no write), --date YYYY-MM-DD (deterministic in-report date), --root PATH. Self-check: --selftest runs the per-rule golden fixtures (adversarial slug corpus, W5 baseline, scope exclusions, output containment, secret-scanner coexistence).
Read the report it wrote (docs/wasted-code/report.md) — per-rule findings (W1 dead-end refs, W2 broken links, W3 missing helpers, W4 footgun config), each labelled broken/confirmed (provable from repo state) or candidate/inconclusive (needs human judgment), plus the Coverage section (W5 swallowed-error baseline, unresolved references, skipped inputs). Silence is auditable, not proof of cleanliness.
Interpret every finding as a review CANDIDATE, never a verdict or an auto-fix — check it against the false-positive classes (flow-skills/find-wasted-code/references/false-positive-examples.md). Ambiguous references live under Coverage → Unresolved, never as defects. The W5 baseline is a count for review, not a finding list; annotate an intentional swallow with find-wasted-code: ignore W5 — <reason>.
Output a Mode A summary in chat: per-rule totals by severity, the W5 baseline counts, and the top confirmed broken items with their file:line. Do not paste raw file contents. This audit is read-only to the project — it writes only the tracked docs/wasted-code/report.md (containment-checked, sentinel-guarded, redaction-safe so it never trips the pre-commit secret scan). Fixes are the operator/PO's call.
For a confirmed broken finding, recommend the concrete fix the report suggests (fix the path, restore the file/handler, or correct the anchor). Point token/ceremony questions at the sibling audits: /token-waste-audit (tokens) and /find-wasted-effort (process ceremony).

Non-Claude surfaces (Codex/Cursor/Copilot/Gemini): invoke the find-wasted-code skill by name and run the analyzer directly (python hooks/local/find-wasted-code.py); it is stdlib-only and surface-independent.