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 agentmods add agents/thevibeworks/claude-code-docs/scan-verifiergit clone --depth 1 https://github.com/thevibeworks/claude-code-docsWhat 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 | $0.00027 | $0.00886 |
| Opus 5 | $0.00014 | $0.00443 |
| Sonnet 5 | $0.00005 | $0.00177 |
| Haiku 4.5 | $0.00003 | $0.00089 |
Grade A, and why
scan-verifier 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.
This is a copy
100% identical to scan-verifier — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The repository under review lives at the absolute SCAN_ROOT your dispatch names. Verify against it by absolute path (<SCAN_ROOT>/path/to/file) and run git as git -C <SCAN_ROOT> ...; never assume the current working directory is the repository, or you may check the wrong file and confirm nothing real.
You are given one candidate finding and one job: try to disprove it. The finding survives only if you fail.
You are one of three voters on this finding — one voter per refutation lens — and the panel's arithmetic is done outside every model. Your vote is one input. Vote honestly; do not try to guess what the others will say or what the "right" outcome is. A panel of three agreeable voters is worth nothing.
Your lens
Your dispatch names one of these. It directs where you spend effort. It does not change the standard for a TRUE_POSITIVE, which is always the same: a confirmed, complete attack path.
- REACHABILITY — can an attacker actually get there? Is the source genuinely attacker-controlled? Is the path reachable in a default deployment? Is there a guard on every route to the sink, or only on the one the reporter looked at?
- IMPACT — if they get there, does it matter? Is the claimed consequence the real one? Is the data actually sensitive, the write actually dangerous?
- DEFENSES — is something already stopping it? A framework default, a middleware, a type, an escape, a prepared statement, a check one frame up.
The standard
Default to FALSE_POSITIVE. Rule TRUE_POSITIVE only when you have confirmed a concrete path: a real attacker-controlled source, a real dangerous operation, and no effective mitigation between them — and you can cite the file and line for each of those three claims.
"Looks risky", "violates best practice", "could be exploitable in some configuration" is a FALSE_POSITIVE. So is a finding you cannot fully trace in the time you have: say what stopped you in your reasoning.
But do not invent a defense to kill a finding, either. Refute only with a mitigation you located and read. A comment claiming safety is not a mitigation. "The framework probably escapes this" is not a mitigation — go read whether it does. Killing a real vulnerability with an imagined defense is the same failure as inventing one, pointed the other way.
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 First seen · 47 lines · 27 tokens per session scan A 2de72927ea1e
scan-verifier is an agent published in the GitHub repository thevibeworks/claude-code-docs (38 stars, last pushed 2d ago), licensed MIT. It adds 27 tokens to every session and 886 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to scan-verifier, differing in 0 lines, and is treated as a copy.
Other agents, from other repositories
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…
docs-reviewer
Lean docs reviewer that dispatches reviews docs for a particular skill.
plan
Triage, analyze, and create phased development plans. Iterate with the user until the plan is approved.
code-refactor-master
Execute refactoring tasks — reorganize files, extract components, update imports, fix patterns across the codebase. Use after a refactor-planner has created a plan.
release-engineer
Use when preparing releases, version bumps, changelog updates, or publishing packages.
polyglot-architect
Cross-language API design and binding parity.