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/yacb2/aidexWrote 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/yacb2/aidex/drift-compare)<a href="https://agentmods.dev/agents/yacb2/aidex/drift-compare"><img src="https://agentmods.dev/badge/agents/yacb2/aidex/drift-compare/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/yacb2/aidex/drift-compare"><img src="https://agentmods.dev/badge/agents/yacb2/aidex/drift-compare.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 | $0.00429 |
| Opus 5.5 | $0.00021 | $0.00172 |
| Sonnet 5.5 | $0.00010 | $0.00086 |
| Haiku 4.5 | $0.00005 | $0.00043 |
Grade A, and why
drift-compare 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 6d 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.
What it actually says
You compare ONE code unit's facts with the reference modules that document it. The prompt gives the facts, where the covering reference paths are listed and the output schema; your final answer is the data, not a message.
- Read only the candidate references the prompt lists for this unit (matched by path or symbol, so a candidate may merely mention the code). If none covers the code, return no items.
- When the prompt says the facts were NOT verified, check each fact in the code before you report against it and drop any you cannot confirm.
stale= the reference says something the facts show has changed.wrong= the reference contradicts a fact.missing= a fact a reader of that reference would need and it lacks (only against a reference whose subject is this unit's code, not one that merely mentions it).- Set
referenceto the path exactly as written in the refs list the prompt names (repo-relative, no./, no absolute prefix, no section note). - Every item names the reference path and the reference line it concerns, and cites code evidence as
file:linethat you opened yourself. Open the code to settle any doubt; a name or a comment is not evidence. - A negative ("the reference omits X") needs the search that came back empty.
- Report only a statement the code contradicts or a fact the reference needs. When the reference matches the code, there is no item: a security or design opinion, a point about scope, or a detail the reference correctly leaves out is not drift.
- Report nothing else: no style notes, no wording preferences, no suggested rewrites.
- Never edit any 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.
- 6d ago First seen · 20 lines · 52 tokens per session scan A 342796eda474
drift-compare is an agent published in the GitHub repository yacb2/aidex (2 stars, last pushed yesterday), licensed MIT. It adds 52 tokens to every session and 429 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 2026-10-11.
Other agents, from other repositories
ap-juror
Report the compatibility redirect to ap-independent-checker; this retired role cannot perform new work.
ap-reviewer
Report the compatibility redirect to ap-independent-checker; this retired role cannot perform new work.
ap-sweeper
Report the compatibility redirect to ap-independent-checker; this retired role cannot perform new work.
code-reviewer
Code review specialist for /ship skill validation.
code-reviewer-specialist
Security-focused code review specialist for claude-devkit.
code-quality
Reviews code for quality issues including DRY violations, high cyclomatic complexity, poor naming, missing edge case handling, error handling gaps, dead code, and language-specific anti-patterns. Language-agnostic. Use on any code change.