ericrisco/rsc-harness / skills/error-handling
error-handling
VerifiedUse when designing the reaction to a class of failures — typed error taxonomies, retry/backoff/timeout policy, circuit breakers, React/Next error boundaries, and the user-message vs operator-log split. NOT diagnosing one specific crash (that is debug), NOT logs/metrics/traces (that is observability), NOT the wire error envelope (that is api-design).
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 12.5 KB |
| evals/README.md | markdown | 820 B |
| evals/cases.yaml | config | 4.3 KB |
| references/boundaries-and-messaging.md | markdown | 4.8 KB |
| references/retry-and-resilience.md | markdown | 5.2 KB |
| scripts/verify.shexecutable | script | 4.2 KB |
scripts/verify.sh
View on GitHub#!/usr/bin/env bash
set -euo pipefail
# ============================================================================
# NAME
# verify.sh — error-handling anti-pattern scanner
#
# USAGE
# ./verify.sh [PATH] # default PATH = .
# ./verify.sh --strict [PATH] # exit non-zero if anything is flagged
#
# WHAT IT DOES
# Greps the working tree for the highest-signal anti-patterns this skill
# teaches and prints file:line for each hit:
# - empty `catch {}` / `catch (e) {}` (swallowed failure)
# - Python `except: pass` and bare `except:` (swallowed / over-broad)
# - `alert(` as a handler, lone `console.log(err)` / `console.error(e)`
# - retry loops with no cap: `while (true)`/`while True` near "retry"
#
# POSTURE
# - READ-ONLY. Never writes, never auto-fixes.
# - HEURISTIC and ADVISORY. It flags candidates; a human judges each one.
# - Exits 0 by default even with findings (warnings only). Pass --strict to
# make any finding exit non-zero (for CI). Exits 0 cleanly on an empty or
# clean target — no false failure.
# - Portable to stock macOS bash 3.2 (no mapfile, no associative arrays).
#
# EXIT CODES
# 0 no findings, OR findings in advisory (default) mode
# 1 findings AND --strict was passed
# ============================================================================
RED=$'\033[31m'; YEL=$'\033[33m'; GRN=$'\033[32m'; RST=$'\033[0m'
if [ -n "${NO_COLOR:-}" ]; then RED=""; YEL=""; GRN=""; RST=""; fi
STRICT=0
TARGET="."
for arg in "$@"; do
case "$arg" in
--strict) STRICT=1 ;;
*) TARGET="$arg" ;;
esac
done
if [ ! -e "$TARGET" ]; then
printf '%s[skip]%s target not found: %s\n' "$YEL" "$RST" "$TARGET" >&2
exit 0
fi
HITS=0
# Prefer ripgrep (respects .gitignore, fast); fall back to grep -r.
if command -v rg >/dev/null 2>&1; then
search() { rg --line-number --no-heading --color=never -e "$1" "$TARGET" 2>/dev/null || true; }
searchP() { rg --line-number --no-heading --color=never -P -e "$1" "$TARGET" 2>/dev/null || true; }
else
search() { grep -rEn --exclude-dir=.git --exclude-dir=node_modules -- "$1" "$TARGET" 2>/dev/null || true; }
searchP() { grep -rEn --exclude-dir=.git --exclude-dir=node_modules -- "$1" "$TARGET" 2>/dev/null || true; }
fi
# report <label> <matches> : print a section if there are matches, bump HITS.
report() {
label="$1"; matches="$2"
if [ -n "$matches" ]; then
count=$(printf '%s\n' "$matches" | grep -c . || true)
printf '\n%s[flag]%s %s (%s)\n' "$YEL" "$RST" "$label" "$count"
printf '%s\n' "$matches"
HITS=$((HITS + count))
fi
}
# 1. Empty catch blocks: `catch {}`, `catch (e) {}`, `catch(err){ }`
report "empty catch block — swallows the failure" \
"$(searchP 'catch\s*(\([^)]*\))?\s*\{\s*\}')"
# 2. Python `except: pass` (allow whitespace / one-liners)
report "Python 'except ...: pass' — swallows the failure" \
"$(searchP 'except[^\n:]*:\s*pass\b')"
# 3. Python bare `except:` (over-broad; catches SystemExit/KeyboardInterrupt)
report "Python bare 'except:' — over-broad, catch the specific type" \
"$(searchP 'except\s*:')"
# 4. alert() used as an error handler (leaks internals, blocks UI)
report "alert() as a handler — leaks internals, no recovery" \
"$(searchP 'alert\s*\(\s*(e|err|error)(\.message)?\s*\)')"
# 5. console.log/error of the error as the ONLY handling (log-and-forget)
report "lone console.log/error(err) — log without classify/surface" \
"$(searchP 'console\.(log|error)\s*\(\s*(e|err|error)\s*\)\s*;?\s*$')"
# 6. Uncapped retry loops: while(true)/while True within a couple lines of 'retry'
report "possible uncapped retry loop — needs an attempt cap + jitter" \
"$(searchP 'while\s*\(\s*true\s*\)|while\s+True\s*:' | grep -iE 'retry|attempt|backoff' || true)"
# ----------------------------------------------------------------------------
printf '\n=== Summary ===\n'
if [ "$HITS" -eq 0 ]; then
printf '%s[ok]%s no anti-patterns flagged under %s\n' "$GRN" "$RST" "$TARGET"
exit 0
fi
printf '%s%s candidate(s) flagged%s — heuristic; review each against ../SKILL.md.\n' \
"$YEL" "$HITS" "$RST"
if [ "$STRICT" -eq 1 ]; then
printf '%s--strict: failing.%s\n' "$RED" "$RST" >&2
exit 1
fi
printf 'advisory mode (exit 0); pass --strict to gate.\n'
exit 0
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the work | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation code | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy. | Stars: 197k | Updated | Claude CodeCodex | Verified | Install |