Deterministic Evidence Oracle (Zero Hallucination)

Example session with this skill installed

Verify the root cause of an authentication failure in auth_service.py using strict line citations and an empirical oracle.

Read your context and instructions
Compiled the deterministic evidence oracle

[EVIDENCE-ORACLE] Verification Dossier

Assertion: Null state token unhandled in OAuth callback
Locus: auth_service.py#L44-L52
Verbatim: state = request.args.get('state')
Oracle Result: beta(s) = 1.0 (Forensically Verified)
Status: VERIFIED

Connects securely to your tools. The creator never sees your data.

About this skill

The Deterministic Evidence Oracle is a forensic grounding and verification framework designed to eradicate model hallucination across complex technical, analytical, and coding tasks.

It halts common agent hallucinations: speculative bug diagnosis, phantom API methods, invented documentation parameters, false verification claims, and ungrounded data assertions.

Key Capabilities:

Exact Line-Range Citation Contract: Enforces clickable markdown links (file:///path#L10-L25) and verbatim grounding snips for every technical claim.
Hard Refusal Gate: Prohibits speculative guessing; triggers immediate standardized refusal notices when context is missing.
Mathematical Verification Oracle (beta(s) = 1): Demands independent empirical verification (exit code 0 execution) before an agent may declare a task complete.
Anti-Hallucination Traps Sentry: Detects phantom library methods, drifted configuration keys, and ghost database tables.

Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Productivity.

Does not bypass physical limitations of missing files. Requires access to the codebase or documentation artifacts to extract verifiable citations.

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 buying a skill.

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.

Frequently Asked Questions

No. All evidence grounding, line span checks, and verification scripts execute strictly locally on your machine.

Yes. It can enforce strict line citation contracts and hard refusal triggers on any codebase, technical spec, or analytical report.

Deterministic Evidence Oracle (Zero Hallucination)