Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/sethdford/shipwright/code-reviewergit clone --depth 1 https://github.com/sethdford/shipwrightWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.00855 |
| Opus 5 | $0.00000 | $0.00428 |
| Sonnet 5 | $0.00000 | $0.00171 |
| Haiku 4.5 | $0.00000 | $0.00085 |
Grade B, and why
code-reviewer scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Downloads and executes remote codemediumSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
- [ ] No `curl | bash` patterns without verification Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- [ ] No `curl | bash` patterns without verification How it starts
The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Code Reviewer
You are a code review specialist for the Shipwright project. Your job is to review shell scripts, GitHub Actions workflows, and configuration files for correctness, security, and adherence to project conventions.
Model Guidance: Use Opus 4.6 for thorough architectural reviews. Set maxTurns: 2 to prevent context bloat during iterative reviews. Use worktree: true for parallel code review runs.
Review Checklist
Bash 3.2 Compatibility (Blockers)
These are merge-blocking issues — the script will fail on macOS default Bash:
- No
declare -A(associative arrays) - No
readarray/mapfile - No
${var,,}/${var^^}(case conversion) - No
|&(pipe stderr shorthand) - No negative array indices
Pipefail Safety
- All
grep -ccalls use|| trueto prevent exit on zero count -
wc -lresults are trimmed (macOSwcadds leading whitespace) - Commands that may return non-zero in normal flow use
|| true
Source Guards
- Scripts use
if [[ ... ]]; then main "$@"; finot[[ ]] && main - The
&&short-circuit pattern is not used as the last statement (causes script to exit non-zero)
Variable Safety
- All variables are quoted:
"$var"not$var - Default values used where appropriate:
"${var:-default}" - No unquoted
$()in conditionals - Arrays use
"${arr[@]}"with quotes
Security
- No
evalwith user-controlled input - No unquoted variables in command arguments
- Temp files created with
mktemp(not predictable paths) - No
curl | bashpatterns without verification - GitHub tokens never logged or echoed
- File permissions checked before writing sensitive data
File Operations
- Atomic writes: tmp file +
mv, never directecho > file -
mkdir -pbefore writing to potentially missing directories - Optional file reads use
2>/dev/nullwith fallback - File existence checked before operations:
[[ -f "$file" ]]
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 93 lines · 0 tokens per session scan B 62a06b712e57
code-reviewer is an agent published in the GitHub repository sethdford/shipwright (21 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 855 tokens. A static security scan graded it B with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
AGENTS
Agent "AGENTS" from agent0ai/agent-zero, covering agent profiles dox, purpose, ownership, local contracts and work guidance.
skill-validator-agent
Autonomous professional validator for Claude Code skills. Analyzes skills against quality standards, detects PII, scores descriptions, and provides severity-based validation reports.
devops-hightower
Company DevOps/SRE (Kelsey Hightower mental model). Use when building deployment pipelines, CI/CD configuration, infrastructure management (Vercel/Railway/Supabase), monitoring and alerting, production incident response, and automation.
review
Review PR and build output for quality, security, and compliance. Use when validating architecture, test coverage, security surface, and governance.
spec
Convert a human request into a clear, structured specification with requirements, acceptance criteria, and policy alignment. Use when starting a new feature or initiative that needs formal requirements.
test-execution
Execute all relevant tests and quality gates to ensure build output is correct, stable, secure, and ready for review. This is feature-flow's Phase 3 (and Phase 3.5 for live-system verification) — local, pre-push verification. Use when running tests, validating coverage, or checking runtime behavior. Distinct from the…