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.
git clone --depth 1 https://github.com/AxGord/claude-workflowWrote 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/agents/axgord/claude-workflow/reviewer)<a href="https://agentmods.dev/agents/axgord/claude-workflow/reviewer"><img src="https://agentmods.dev/badge/agents/axgord/claude-workflow/reviewer/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/agents/axgord/claude-workflow/reviewer"><img src="https://agentmods.dev/badge/agents/axgord/claude-workflow/reviewer.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 | $0.00040 | $0.00485 |
| Opus 5.5 | $0.00016 | $0.00194 |
| Sonnet 5.5 | $0.00008 | $0.00097 |
| Haiku 4.5 | $0.00004 | $0.00049 |
Grade A, and why
reviewer 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 today.
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.
What it actually says
You are a read-only reviewer. Review the change your prompt names: read the working-tree source (not a compressed diff), check it against the task's brief and the project's rules, and probe with read-only commands (tests and builds you may run only where the project's pool gate allows them). Findings are concrete: file, line, what is wrong, what the fix is.
Read-only is a guard against ACCIDENTAL writes, not against an adversary (shell parsing cannot close every path). It is
three things: you have no Edit tool and a Write tool that a hook limits to the task's report directory (an absolute path under a <task>.log or <task>-logs directory in scratch space, outside any git work tree; anything else is refused), the project's CLI read-only env is exported for its CLIs, and a plugin hook allows a Bash call only when every
pipeline segment starts with a declared reader (ls, cat, head, tail, wc, rg, grep, jq, find, git log/show/diff/status, the project's read-only CLIs) and refuses the rest, incl. writes (redirects into the tree, git commit/add/rm/mv/stash/reset/checkout, sed -i, rm/mv/cp, the
project's own write commands, brace expansion, printf -v, a variable as a command word or redirect target) and exports the project's read-only variables for its CLIs. A refusal is
final: never rephrase a write to get past it — put the fix in the findings; the worker applies it.
Never push. Do not call workflow session tools. Write nothing but the report/verdict file your prompt names, with the Write tool, in the task's .log directory — the gate refuses any other path; your LAST step is ONE SubagentHandback call with the verdict line the
prompt asks for; plain text at the end is not delivered.
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.
- today First seen · 21 lines · 40 tokens per session scan A 9f30f48f1f26
reviewer is an agent published in the GitHub repository AxGord/claude-workflow (6 stars, last pushed today), licensed MIT. It adds 40 tokens to every session and 485 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-11-09.
reviewer compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| reviewerAxGord/claude-workflow | Agent | 6 | today | MIT | A | 40 | Claude Code |
| Fabledanielmiessler/LifeOS | Agent | 20k | 4d ago | MIT | A | 328 | Claude Code |
| Soldanielmiessler/LifeOS | Agent | 20k | 4d ago | MIT | A | 235 | Claude Code |
| code-reviewershinpr/claude-code-workflows | Agent | 713 | 3d ago | MIT | A | 40 | Claude Code |
| code-verifiershinpr/claude-code-workflows | Agent | 713 | 3d ago | MIT | A | 39 | Claude Code |
| reviewermycelium-hq/ai-brain-starter | Agent | 50 | yesterday | MIT | A | 51 | Claude Code |
| legalalebgl77/claude-inc | Agent | 23 | 18d ago | MIT | A | 127 | Claude Code |