Request Guard False-Positive Investigator
Compare a flagged request with a control and investigate a possible false positive without automatically changing policy.
$2
Works with the AI Agents you already use
Request Guard False-Positive Investigator
Example session with this skill installed
Use $request-guard-false-positive-investigator to complete this task.
Investigate whether flagged versus control in input.json supports a false-positive finding. Both examples are synthetic; there is no independent evidence of legitimacy. Explain exactly what differs, the observed result and unresolved ground truth. Do not change policy.
The trusted local release binary is /absolute/path/request-guard-mcp It is the verified Rust v1.0.0 binary. You may execute the packaged helper if needed. Save a concise human-facing report as acceptance-report.md and state actual evidence and limitations. Do not modify the skill or scripts. All inputs are synthetic. Keep everything private and unpublished.
- Read your context and instructions
- Compiled the request guard false-positive
Recorded acceptance-run output from Codex CLI 0.168.0 on Linux, November 13, 2026, using package 0.1.1 and released Rust server v1.0.0. All request fixtures were synthetic. Local filesystem paths are redacted; this is recorded output, not a live demo.
False-positive investigation
Finding: a false positive is not established. The pair demonstrates sensitivity to one user-agent field. Both requests are synthetic, and neither has independent evidence of legitimacy. Ground truth remains unknown, as explicitly stated by the user; the name “control” does not establish legitimate traffic.
Exact input comparison
Only user_agent differs: flagged is sqlmap; control is Mozilla/5.0. Both have method GET, path /products/configurable-shirt, headers.accept = text/html, and headers.accept-language = en. No other input fields differ. The helper sends both header values as present; it assesses header presence rather than their original values. The path needs no query/fragment removal.
Observed results
| Output | Flagged (synthetic) | Control (synthetic) |
|---|---|---|
| Verdict | block | allow |
| Confidence | medium | very_low |
| Score | 1 | 0 |
| Threat category | scraping | none |
| Signals | ua_scraper: value 1, weight 0.7 | none |
| Recomputed explanation factors | ua_scraper: contribution 100% | none |
The reported signal description is “User-agent matches scraping tool pattern.” The explanation's 100% contribution is distinct from the raw signal weight of 0.7. The single-field comparison isolates the user-agent change within this fixture: substituting Mozilla/5.0 removes the reported signal and changes the verdict. This is consistent with a user-agent pattern match; it does not establish that blocking the original was wrong. User agents are self-declared. The control's allow result and the tool's “appears legitimate” recommendation do not verify legitimacy; the block result does not prove malicious intent.
Evidence and limitations
Captured 2026-11-13 at approximately 07:08:40 UTC using the packaged helper in false-positive mode, with client identity request-guard-paid-workflow version 0.1.0 (skill package 0.1.1). The user supplied the trusted, verified Rust binary at /private-test-workspace. This run checked reported server identity request-guard-mcp version 1.0.0; it did not independently verify a signature. The helper discovered 23 tools and successfully called classify and explain for both items over local stdio. Raw normalized inputs and results are retained in false-positive-evidence.json. No item errors were reported.
Explanations were recomputed from supplied metadata, not retrieved from stored decisions. Optional backends were unconfigured. This is one local synthetic comparison, with no observed traffic, authenticated identity, behavioral history, authorization evidence, live enforcement, or integration testing. Request IDs and latency are incidental and are not policy-change evidence or benchmark results.
Proposed human review
For this site's /products/configurable-shirt route, first obtain independently verified authorized traffic and assess whether legitimate clients actually use the matched user-agent. If confirmed, review a narrowly scoped reduction in the influence of ua_scraper for authenticated, authorized clients on that route, with abuse regression checks before adoption. The present evidence does not justify an adjustment or a user-agent-only bypass. No policy, allowlist, firewall, DNS, credentials, or deployment settings were changed; nothing was published.
Connects securely to your tools. The creator never sees your data.
About this skill
Identify exactly which metadata fields differ, compare verdicts and signals, and distinguish a synthetic counterfactual from independently verified legitimate traffic. A changed verdict does not establish that the original was wrong.
Included
Versioned SKILL.md instructions, a standalone Node.js helper, report guidance, compatibility metadata and the GPL-3.0-or-later license. Sample inputs are included where the workflow uses them. Package version: 0.1.1. One-time price: US $2.
Requirements and verification
Use a compatible SKILL.md agent and Node.js 18+. For live local assessments, separately install and checksum/attestation-verify the Rust Request Guard MCP v1.0.0 binary. The helper checks the reported version, not the binary signature. Windows x64 and Linux x64 glibc helper checks are recorded; these exact ZIPs also passed controlled Codex CLI 0.168.0 acceptance on Linux over stdio.
Limits
Request Guard assesses metadata; it does not intercept traffic or enforce policy. User agents are self-declared. Allow is not a safety guarantee; block is not proof of malicious intent. Do not send cookies, authorization headers or bodies. No certification of other agent UIs, remote transports, optional backends or live defense integrations is claimed. Cloudflare fixtures and policy differences in the demo are synthetic. No automatic firewall, DNS or enforcement changes are made.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 3 days ago
Requires
Permissions
Reads selected sanitized inputs and writes local reports. Local helpers use process-local stdio and clear optional-backend environment configuration. No credentials, firewall, DNS or enforcement changes are needed.