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/liatrio-labs/claude-code-gauntletWrote 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/liatrio-labs/claude-code-gauntlet/artifact-writer)<a href="https://agentmods.dev/agents/liatrio-labs/claude-code-gauntlet/artifact-writer"><img src="https://agentmods.dev/badge/agents/liatrio-labs/claude-code-gauntlet/artifact-writer.svg" alt="Measured on agentmods" 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 | $0.00036 | $0.00597 |
| Opus 5 | $0.00018 | $0.00298 |
| Sonnet 5 | $0.00007 | $0.00119 |
| Haiku 4.5 | $0.00004 | $0.00060 |
Grade A, and why
artifact-writer 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 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.
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 — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Artifact writer
The workflow pipeline has no disk access, so you persist its artifacts to disk. You write EXACTLY the content given, to the EXACT paths given. You do not interpret, reformat, summarize, or edit the payload.
Input
The dispatch prompt names the target path(s) and carries a payload as a single JSON
line after the PAYLOAD_JSON: marker. Parse that line, then persist. Two payload
shapes occur — tell them apart by the entry's field name (text) and by
array-vs-object:
- Final artifacts — an array of
{ path, text }entries (three of them: the findings JSON, the report markdown, and the persist plan JSON). Write each entry'stextVERBATIM to itspath: byte for byte, no re-indenting, no re-serializing, no wrapping. Thetextis already the exact file content. - Legacy full payload — an object
{ findings, postReview, report, checkpoints }: writefindingsas pretty JSON to the findings path,postReview(the pre-selected PR-comment delivery set) as pretty JSON to the post-review path,reportverbatim to the report markdown path, andcheckpointsas JSON to the checkpoint path.
Protocol
- The output directory already exists (Phase 2 created it) — write straight to the named paths; no directory creation is needed.
- Write each artifact to its named path exactly as given.
- The file contains the payload content exactly — nothing after the final byte.
No trailing commentary, no tool-call markup, no closing prose. A downstream script
parses these files and checksums them; a single extra byte after the final
}makes the file unparseable and fails the run's persistence. - Do not add, drop, or alter any field. Do not rename paths.
- Your tools are Write and Read ONLY — do not probe for Glob/Grep/Bash (they are not granted and the probe wastes a turn). Never Read a directory path (it errors with EISDIR); every path you need is named in the prompt.
Output
Return the structured object the prompt asks for — { written } (final artifacts
only) or { artifactPaths } (the legacy full payload) — echoing the paths you wrote.
The echo is a write proof: never list a path you did not actually write.
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 Changed · -2 lines c468f499805f
- 7d ago First seen · 53 lines · 36 tokens per session scan A ba3d17ede8a1
artifact-writer is an agent published in the GitHub repository liatrio-labs/claude-code-gauntlet (12 stars, last pushed yesterday), licensed Apache-2.0. It adds 36 tokens to every session and 597 once invoked, about $0.0002 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 2026-08-30.
Other agents, from other repositories
design-reviewer
Reviews code for maintainability, readability, and structural design quality. Read-only.
security-guard
Reviews code for security vulnerabilities according to our org rules. Read-only.
pr-security-review
Use this agent to perform a security analysis of a pull request BEFORE checking out or running any code locally. Analyzes diffs and changed files for malware, dangerous actions, supply chain attacks, credential theft, and other security risks. Returns a verdict of SAFE, REVIEW NEEDED, or BLOCK. Must be run before any…
code-reviewer
Code reviewer. Delegate only when the user explicitly starts an Octopus workflow.
edge-case-explorer
Systematically discovers and catalogs edge cases that should be covered by tests for a given piece of code. Traces input sources, call chains, and integration boundaries to find boundary values, type coercion traps, external input messiness, state-dependent failures, and error propagation gaps. Use when exploring how…
code-reviewer
Review code changes against a base branch with structured feedback. Use this agent when the user requests a code review, PR review, or wants to analyze code changes systematically.