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 skills add wan-huiyan/agent-traffic-control --skill code-reviewer-subagent-no-bash-blocked-on-pr-diffgit clone --depth 1 https://github.com/wan-huiyan/agent-traffic-controlWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/wan-huiyan/agent-traffic-control/code-reviewer-subagent-no-bash-blocked-on-pr-diff)<a href="https://agentmods.dev/skills/wan-huiyan/agent-traffic-control/code-reviewer-subagent-no-bash-blocked-on-pr-diff"><img src="https://agentmods.dev/badge/skills/wan-huiyan/agent-traffic-control/code-reviewer-subagent-no-bash-blocked-on-pr-diff/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/wan-huiyan/agent-traffic-control/code-reviewer-subagent-no-bash-blocked-on-pr-diff"><img src="https://agentmods.dev/badge/skills/wan-huiyan/agent-traffic-control/code-reviewer-subagent-no-bash-blocked-on-pr-diff.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00214 | $0.01780 |
| Opus 5 | $0.00107 | $0.00890 |
| Sonnet 5 | $0.00043 | $0.00356 |
| Haiku 4.5 | $0.00021 | $0.00178 |
Grade A, and why
code-reviewer-subagent-no-bash-blocked-on-pr-diff scanned grade A with 0 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 10d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Code-reviewer subagents have no Bash — blocked when told to gh pr diff
Problem
You dispatch a review-panel subagent (e.g. feature-dev:code-reviewer) to review
open GitHub PRs, with a prompt like "use gh pr diff <N> to get each PR's diff,
then review." The agent returns a BLOCKED report — no findings — explaining it
has no shell/gh/git tool and the PR source isn't in the working tree. Meanwhile
a sibling reviewer in the same panel may succeed (if it happened to find a
materialized worktree on disk). The blocked run is wasted compute and, worse, can
masquerade as "clean" if you don't read it carefully.
Context / Trigger Conditions
- Dispatching
feature-dev:code-reviewer,voltagent-qa-sec:*,Explore, or similar review/search agents against a GitHub PR or a branch not checked out in the main working tree. - The agent's report says any of: "No shell /
gh/gittool is exposed", "The PR sources are not in the working tree", "WebFetch returns 404 (private repo)", or it reviewed files on the current branch (oftenmain, which predates the PR) instead of the PR's changes. - In a parallel panel, asymmetric outcomes — some reviewers found the code, one came back blocked — because the successful ones discovered an existing worktree path and the blocked one looked in the main checkout.
Root cause
These agents' toolsets exclude Bash. For example feature-dev:code-reviewer has
Glob, Grep, LS, Read, NotebookRead, WebFetch, TodoWrite, WebSearch, KillShell, BashOutput — note KillShell/BashOutput are present but Bash itself is
not, so it can read a background shell's output but cannot start a command.
No Bash → no gh pr diff, no git checkout <pr-branch>, no git diff. Read
hits the filesystem directly, so the agent can only see whatever branch is
currently checked out in the working tree it lands in (the main repo checkout,
which predates the PR). Private repos also defeat the WebFetch fallback (404).
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.
- 10d ago First seen · 127 lines · 214 tokens per session scan A 611f580af2f2
code-reviewer-subagent-no-bash-blocked-on-pr-diff is a skill published in the GitHub repository wan-huiyan/agent-traffic-control (3 stars, last pushed 4d ago), licensed MIT. It adds 214 tokens to every session and 1,780 once invoked, about $0.0011 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
skeptical-triage
Reusable 3-round self-challenge + arbiter pattern for filtering false positives from findings/verdicts. Use when the cost of a false-positive gate block exceeds the cost of 4 extra LLM turns.
vertical-real-estate
Residential-proptech domain knowledge so architect / pm aren't naive when speccing real-estate products (listings, lead-crm, transaction-coordination, property-mgmt). Codifies MLS/IDX reality, listing status lifecycle + syndication canonical-source, long-cycle lead nurture, transaction-coordination as the high-pain…
well-architected
6-pillar architecture review framework. Adapted from AWS Well-Architected for use by greatcto's architect agent on every non-nano ARCH document. Forces explicit answers across operational excellence, security, reliability, performance, cost, and sustainability — not just feature design.
product-economics
Does this product make money at a price someone will pay? Forces contribution margin, a price with a stated basis, and a bottom-up market size — each number labelled measured / assumed / unknown, so a guess can never be read as a calculation.
pm-planning
Decomposition methodology for pm agent — turns an approved ARCH document into a Beads task list with explicit dependencies, time-boxes, and acceptance criteria. The pipeline can only orchestrate work it can see; this skill defines what "seeable work" looks like.
decision-eval
Spawns the decision-scorer agent after architect proposes 2+ variants in an ADR. Produces a weighted scoring table and recommended choice saved to docs/decisions/.