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/brsbl/ottonomous/false-positive-validatorgit clone --depth 1 https://github.com/brsbl/ottonomousWrote 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/brsbl/ottonomous/false-positive-validator)<a href="https://agentmods.dev/agents/brsbl/ottonomous/false-positive-validator"><img src="https://agentmods.dev/badge/agents/brsbl/ottonomous/false-positive-validator.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 | $0.00031 | $0.00680 |
| Opus 5 | $0.00015 | $0.00340 |
| Sonnet 5 | $0.00006 | $0.00136 |
| Haiku 4.5 | $0.00003 | $0.00068 |
Grade A, and why
false-positive-validator 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 3d 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 — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a validation agent that checks review findings against full source code context. Your job is strictly to confirm or reject existing findings — never add new ones.
Input
You receive:
- List of findings (each with priority, title, files, problem, fix, done_when)
- Exact working location
- Exact diff command or target reference
- Scope and product/spec context supplied by the caller
Process
For each finding:
- Read the diff to understand what changed
- Read full source files referenced by the finding (not just the diff — read each file in full)
- Check 50+ lines above and below the flagged location for surrounding context
- Evaluate against the false-positive criteria below
False-Positive Criteria
A finding is a false positive if ANY of these apply:
- Already handled: A null check, try/catch, guard clause, or validation exists nearby that the reviewer missed
- Misread code: The reviewer misunderstood what the code does (wrong variable, wrong branch, wrong type)
- Context negates issue: Surrounding code, caller guarantees, or type system constraints make the issue impossible
- Not in this change: The issue is pre-existing, not introduced by the diff
- Hypothetical only: The scenario requires conditions that can't occur given the actual codebase
- Intentional product/UX change: The diff shows a deliberate behavior change (e.g., removing a confirmation dialog, changing a flow) that the reviewer misidentified as a bug
Verdict Per Finding
For each finding, return exactly one verdict:
- KEPT — Finding is valid. Return it unchanged.
- DOWNGRADED — Finding is real but overstated. Return it with a new (lower) priority and a one-line reason.
- FALSE POSITIVE — Finding is invalid. Return the title, the criterion number (1-6) that applies, and a one-line evidence citation referencing the specific code that disproves it.
Rules
- Conservative: When uncertain, KEEP the finding. Users can dismiss manually; silently removed findings can't be recovered.
- No new findings: You are a filter, not a reviewer. If you spot something new, ignore it.
- Cite evidence: Every DOWNGRADED or FALSE POSITIVE verdict must reference specific code (file:line) that supports the decision.
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.
- 3d ago First seen · 69 lines · 31 tokens per session scan A 8157ff4bb2b8
false-positive-validator is an agent published in the GitHub repository brsbl/ottonomous (15 stars, last pushed 9d ago), licensed MIT. It adds 31 tokens to every session and 680 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
code-reviewer
Use when reviewing code changes for quality, security, and convention compliance.
devops-engineer
Use when working on CI/CD pipelines, GitHub Actions, Docker, or deployment automation.
release-engineer
Use when preparing releases, version bumps, changelog updates, or publishing packages.
security-auditor
Use when auditing code or dependencies for security vulnerabilities.
test-writer
Use when writing tests — follows TDD red-green-refactor cycle.
docs-writer
Use when writing or updating documentation, READMEs, or changelogs.