A system for marking statements as facts, inferences, or guesses and assigning confidence to them.

Use it when writing explanations or reports that need clear evidence levels and calibrated confidence.

It helps readers distinguish verified information from conclusions and uncertainty instead of treating every claim as equally reliable.

provenance — Skill for Claude Code and Codex by Kshitijpalsinghtomar

Install

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.

Any agent

npx skills add Kshitijpalsinghtomar/depth-skills --skill ds-provenance

Clone the repo

git clone --depth 1 https://github.com/Kshitijpalsinghtomar/depth-skills

Made for: Claude Code, Codex.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/kshitijpalsinghtomar/depth-skills/ds-provenance/github.svg)](https://agentmods.dev/skills/kshitijpalsinghtomar/depth-skills/ds-provenance)

Your own site

<a href="https://agentmods.dev/skills/kshitijpalsinghtomar/depth-skills/ds-provenance"><img src="https://agentmods.dev/badge/skills/kshitijpalsinghtomar/depth-skills/ds-provenance/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/kshitijpalsinghtomar/depth-skills/ds-provenance"><img src="https://agentmods.dev/badge/skills/kshitijpalsinghtomar/depth-skills/ds-provenance.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 43 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 4,472 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00043	$0.04472
Opus 5.5	$0.00017	$0.01789
Sonnet 5.5	$0.00009	$0.00894
Haiku 4.5	$0.00004	$0.00447

Measured 1mo ago against content hash c8ee25f6b310, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

provenance 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 1mo 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.

skills/ds-provenance/SKILL.md · 467 lines

How it starts

The opening of the file, as written. The whole thing — 467 lines — stays where its author put it; the contents beside it link to each section on GitHub.

PROVENANCE v2.0 — Evidence Tagger & Confidence Calibrator (Premium)

Mathematical Compliance: This skill implements the DEPTH-MATHEMATICS specification. It computes and outputs ADS phase activation, CEI components, SI components, cognitive traces (calibration curve, epistemic audit trail), and verifies Gates V5, V6. It provides the EC (Epistemic Calibration) metric for the Depth .

You are about to deliver claims. Some you know. Some you inferred. Some you guessed. Your output treats all three identically — same tone, same phrasing, same implicit confidence.

The user cannot tell which is which. A fact and a guess, written in the same authoritative voice, look the same. The user either trusts everything (risky) or questions everything (wasteful). This skill makes the difference visible. Now it quantifies calibration, detects inflation, and produces audit trails.

The Failure Mode You Must Recognize

You are about to write:

"PostgreSQL JSONB columns support GIN indexing, which will give you fast queries and your team will find the migration straightforward."

Three claims, three different evidence levels, one uniform confident voice:

"JSONB supports GIN indexing" — documented fact
"will give you fast queries" — inference (depends on their specific query patterns)
"team will find it straightforward" — pure guess (you don't know the team)

This is epistemic flattening: collapsing facts, inferences, and guesses into a single confident tone. The user makes decisions based on the guess as if it were a fact.

New failure mode in v2.0: Calibration drift — your confidence scores don't match actual accuracy over time. Inflation creep — guesses gradually get tagged as inferences, inferences as facts. Audit gap — no trail to verify tagging decisions.

The Protocol

0 — PARAMETERIZE: Read Thinking Parameters

Read thinking_parameters:

min_fact_ratio (default 0.6) — minimum F/(F+I+G) ratio
max_guess_tolerance (default 0.15) — maximum G/(F+I+G) ratio
require_calibration_curve (default true) — compute calibration curve
require_audit_trail (default true) — produce epistemic audit trail
inflation_detection (default true) — run inflation audit

Read the full file on GitHub · 467 lines

Changes

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.

1mo ago Changed · +269 lines · +18 tokens per session c8ee25f6b310
2mo ago First seen · 198 lines · 25 tokens per session scan A a9ebf676d537

Subscribe to this mod's changes

provenance is a skill published in the GitHub repository Kshitijpalsinghtomar/depth-skills (6 stars, last pushed 24d ago), licensed MIT. It adds 43 tokens to every session and 4,472 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 2026-08-31.

Research agent-skills ai-agents ai-skills ai-tools audit calibration confidence epistemic evidence inflation metrics trust

provenance next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
provenanceKshitijpalsinghtomar/depth-skills	Skill	6	24d ago	MIT	A	43	Claude Code, Codex
bailian-model-recommendmodelstudioai/skills	Skill	59	yesterday	Apache-2.0	A	136	Claude Code, Codex
bailian-train-deploymodelstudioai/skills	Skill	59	yesterday	Apache-2.0	A	321	Claude Code, Codex
vox-video-directormodelstudioai/skills	Skill	59	yesterday	Apache-2.0	A	238	Codex
aspire-project-v2-migrationmicrosoft/aspire-skills	Skill	93	yesterday	MIT	A	0	Claude Code
aspireifymicrosoft/aspire-skills	Skill	93	yesterday	MIT	A	0	Claude Code
ai-provider-mistral-sdkagents-inc/skills	Skill	26	12d ago	MIT	A	45	Claude Code, Codex