audit-claims

audit-claims is a skill for Claude Code from TalonT-Org/AutoSkillit. It costs 45 tokens per session (3,293 once invoked), scanned A, original, MIT.

A research pull-request review tool that extracts claims from each report section and checks them against available evidence. A pull request is a proposed change awaiting review in a code repository.

In plain words
What is it for?
Use it to audit citation integrity, classify unsupported claims, post review comments, and produce an approval verdict.
Why use it?
It helps find claims that citations or other evidence do not support before the pull request is approved.

Skill for Claude Code

Written for Claude Code: hooks in frontmatter. Also seen: mentions subagents.

Part of the autoskillit plugin — 121 skills, 1 MCP server shipped together

Good fit Use it to audit citation integrity, classify unsupported claims, post review comments, and produce an approval verdict.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/talont-org/autoskillit/audit-claims
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 TalonT-Org/AutoSkillit --skill audit-claims
Clone the repo
git clone --depth 1 https://github.com/TalonT-Org/AutoSkillit

Made for: Claude Code.

Or install autoskillit, the plugin that ships this one along with the rest of its 121 skills, 1 MCP server.

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.

agentmods badge for audit-claims

README.md
[![agentmods](https://agentmods.dev/badge/skills/talont-org/autoskillit/audit-claims/github.svg)](https://agentmods.dev/skills/talont-org/autoskillit/audit-claims)
Your own site
<a href="https://agentmods.dev/skills/talont-org/autoskillit/audit-claims"><img src="https://agentmods.dev/badge/skills/talont-org/autoskillit/audit-claims/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.

agentmods 80×15 button for audit-claims

Your own site · 80×15
<a href="https://agentmods.dev/skills/talont-org/autoskillit/audit-claims"><img src="https://agentmods.dev/badge/skills/talont-org/autoskillit/audit-claims.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 45 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,293 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00045 $0.03293
Opus 5 $0.00023 $0.01647
Sonnet 5 $0.00009 $0.00659
Haiku 4.5 $0.00005 $0.00329

Measured 9d ago against content hash 9f74af040552, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

audit-claims 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 9d 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.

src/autoskillit/skills_extended/audit-claims/SKILL.md · 361 lines

How it starts

The opening of the file, as written. The whole thing — 361 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Audit Claims Skill

Perform a two-phase citation integrity audit on an open GitHub research PR using parallel subagents. Phase 1 extracts claims by report section; Phase 2 matches each claim against available evidence and generates findings. Posts inline review comments and emits a verdict for recipe routing.

Arguments

/autoskillit:audit-claims <worktree_path> <base_branch> <pr_url>

  • worktree_path — Absolute path to the research worktree (skill derives feature_branch from git rev-parse --abbrev-ref HEAD inside it)
  • base_branch — The base branch the PR targets (e.g., "main")
  • pr_url — Explicit PR URL passed by the recipe (avoids re-discovering the PR)

When to Use

  • Called by the research recipe orchestrator after review_research_pr
  • Both read-only gates complete before any resolution step begins
  • Can be invoked standalone to audit citation integrity for any open research PR

Critical Constraints

NEVER:

  • Create files outside {{AUTOSKILLIT_TEMP}}/audit-claims/
  • Approve a PR that has changes_requested findings
  • Post review comments when gh is unavailable — output verdict=approved and exit 0
  • Review files outside the PR diff — scope all audit to diff content only
  • Modify any source code
  • Run deterministic diff annotation (claim positions are report-level, not line-level)
  • Generate findings for experimental claims — they are self-evidencing by definition
  • Run subagents in the background (run_in_background: true is prohibited)

ALWAYS:

  • Use the explicit pr_url argument instead of re-discovering via gh pr list
  • Output verdict= on the final line
  • Exit 0 in all normal cases; verdict drives recipe routing via on_result, not exit code
  • Exit non-zero only for unrecoverable errors (e.g., gh CLI truly unavailable after graceful degradation)
  • Use model: "sonnet" when spawning all subagents via the Task tool
  • Deduplicate findings by (file, line) pairs before posting

Workflow

Step 0: Validate Arguments

Read the full file on GitHub · 361 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.

  1. 9d ago First seen · 361 lines · 45 tokens per session scan A 9f74af040552

Subscribe to this mod's changes

audit-claims is a skill published in the GitHub repository TalonT-Org/AutoSkillit (5 stars, last pushed today), licensed MIT. It adds 45 tokens to every session and 3,293 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 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

insight-error-page

Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…

vercel/next.js · 83 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens