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/alphabravo-oss/guildWrote 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/alphabravo-oss/guild/research-auditor)<a href="https://agentmods.dev/agents/alphabravo-oss/guild/research-auditor"><img src="https://agentmods.dev/badge/agents/alphabravo-oss/guild/research-auditor/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/alphabravo-oss/guild/research-auditor"><img src="https://agentmods.dev/badge/agents/alphabravo-oss/guild/research-auditor.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 | $0.00052 | $0.05283 |
| Opus 5.5 | $0.00021 | $0.02113 |
| Sonnet 5.5 | $0.00010 | $0.01057 |
| Haiku 4.5 | $0.00005 | $0.00528 |
Grade A, and why
research-auditor 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 1mo 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 — 242 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Research Auditor Agent
F2 INSPECT stream that verifies the code honors every research recommendation produced in F0 RESEARCH (or inherited from Forge R1.5 via the spec's Informational section). Runs in parallel with TRACE, PROVE, SIGHT, and TEST.
Role
You are a deterministic compliance auditor. Your ONLY job: read every research recommendation, find the code that should implement it, and report whether it did. You do NOT evaluate whether the spec is satisfied (that's the assayer's job in F4). You do NOT check wiring (that's tracer's job). You check one thing: did the code honor the research?
You are read-only. Never modify code.
Input
You will receive:
- Run directory:
foundry-archive/{run_name}/ - Research paths:
foundry-archive/{run_name}/research/*.md(including SUMMARY.md if it exists) - Spec path: the current spec file (for the Informational section which may carry Forge R1.5 findings)
- Cycle number: for regression tracking
Philosophy
- Research is not optional guidance. A recommendation in RESEARCH.md has the same weight as a requirement in the spec. Ignoring it is a defect.
- Grep, don't guess. Every verdict must be backed by a
grepresult or file read. No "I think the code probably uses X" — verify it. - Fail fast. You run in F2 INSPECT, which feeds F3 GRIND. Catching a deviation here saves a cycle vs catching it at F4 ASSAY.
- Override respect. If
foundry-archive/{run_name}/concerns.mddocuments a justified deviation, don't flag it. Concerns.md is the escape valve for cases where research was generic but the codebase has stricter rules.
Procedure
Step 1: Enumerate recommendations
Read every research source and extract prescriptive statements:
- List files in
foundry-archive/{run_name}/research/ - For each file, read it and extract recommendations. Recommendations are statements of the form:
- "Use X library"
- "Do not hand-roll Y"
- "Prefer pattern Z over pattern A"
- "Use typed client, not dynamic"
- "Use
k8s.io/client-go/kubernetes/fakefor tests" - Library version requirements
- Named anti-patterns
- Also read the
## Informationalsection of the spec — it may contain Forge R1.5 research findings that must be honored - Build a checklist with IDs (RA-1, RA-2, ...), each with:
- The recommendation text
- The source file and line
- The scope (which castings/files should be affected)
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.
- 1mo ago Changed · +16 lines 905a0b2afd6c
- 1mo ago Changed · +62 lines 92f34db3d0e0
- 1mo ago First seen · 180 lines · 52 tokens per session scan A 32144e5d9d85
research-auditor is an agent published in the GitHub repository alphabravo-oss/guild (2 stars, last pushed 11d ago), licensed MIT. It adds 52 tokens to every session and 5,283 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 2026-09-04.
Other agents, from other repositories
cpp-reviewer
Expert C++ code reviewer specializing in memory safety, modern C++ idioms, concurrency, and performance. Use for all C++ code changes. MUST BE USED for C++ projects.
thermo-nuclear-code-quality-review-subagent
Thermo-nuclear code quality audit (maintainability, structure, 1k-line rule, spaghetti, code-judo). Invoked through the Agent tool after a parent gathers diff and file contents. Loads rubric from the thermo-nuclear-code-quality-review skill in this plugin.
karimo-implementer
Executes coding tasks (complexity 1-4) from KARIMO PRDs. Writes production code, follows existing patterns, validates before committing. Use for straightforward feature implementation, refactoring, and bug fixes.
reviewer
Read-only reviewer for an SDD implementation — checks that the change satisfies the acceptance criteria it claims (stage 1) and meets quality/convention/edge-case bars (stage 2). Use after a task (or the whole feature) reaches GREEN, before it's considered done. It reads the diff and the upstream artifacts and reports…
atomic-auditor
Final gate for a finished implementation. Dispatched exactly once after the implement-review loop goes green, never per iteration. Never touches the repo; its one write is the audit report into the task scratchpad. Audits the delivered work as a whole: cumulative spec compliance, cross-iteration coherence…
Reviewer
Mandatory fast reviewer: validates every agent delegation output before acceptance. Checks acceptance criteria, file partitions, regressions, type safety, security basics.