truthgate

truthgate is a skill for Claude Code, Codex from ay7627514-dotcom/truthgate. It costs 23 tokens per session (334 once invoked), scanned A, original, MIT.

A verification workflow for Hermes tasks that checks factual claims, multi-step work, and external actions against evidence. Hermes is the task system this skill is designed to govern.

In plain words
What is it for?
Running a preflight check, following the approved plan, collecting tool and test evidence, and submitting the result for verification before reporting completion.
Why use it?
It reduces the risk of claiming that work is complete when the evidence does not support it. High-risk actions can also require a separate, task-specific approval token.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/ay7627514-dotcom/truthgate/hermes
Any agent
npx skills add ay7627514-dotcom/truthgate --skill hermes
Clone the repo
git clone --depth 1 https://github.com/ay7627514-dotcom/truthgate

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.

agentmods badge for truthgate

README.md
[![agentmods](https://agentmods.dev/badge/skills/ay7627514-dotcom/truthgate/hermes.svg)](https://agentmods.dev/skills/ay7627514-dotcom/truthgate/hermes)
Your own site
<a href="https://agentmods.dev/skills/ay7627514-dotcom/truthgate/hermes"><img src="https://agentmods.dev/badge/skills/ay7627514-dotcom/truthgate/hermes.svg" alt="Measured on agentmods" height="20"></a>
Per session 23 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 334 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00023 $0.00334
Opus 5 $0.00012 $0.00167
Sonnet 5 $0.00005 $0.00067
Haiku 4.5 $0.00002 $0.00033

Measured 4d ago against content hash ddc9eb88387e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

truthgate 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 4d 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.

integrations/hermes/SKILL.md · 38 lines

What it actually says

TruthGate for Hermes

Use TruthGate for factual claims, multi-step implementation, external actions, and any task where a false completion claim would cost time, money, or data. Do not add TruthGate overhead to casual conversation or creative drafting.

Required flow

  1. Call mcp_truthgate_truthgate_preflight before using external tools.
  2. Stop if status is blocked or needs_clarification.
  3. Preserve the returned task_id, task_digest, spec, and plan.
  4. Execute only the approved plan through Hermes' existing tools and approval controls.
  5. Collect evidence IDs, tool results, exit codes, test output, and artifact hashes. Tool traces must come from a trusted host wrapper when signed traces are required.
  6. Submit the final candidate and evidence through mcp_truthgate_truthgate_verify.
  7. Report completion only when TruthGate returns verified.

For a high-risk or irreversible action, a human issues a task-bound token with the truthgate approve CLI. Call mcp_truthgate_truthgate_authorize immediately before the action. The token is single-use.

Non-negotiable rules

  • Never fabricate evidence or tool traces.
  • Never reinterpret blocked as permission to continue.
  • Never claim that a file, message, workflow, or deployment changed without direct tool evidence.
  • A failed verification means repair the work or report the exact blocker.
Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 4d ago First seen · 38 lines · 23 tokens per session scan A ddc9eb88387e

Subscribe to this mod's changes

truthgate is a skill published in the GitHub repository ay7627514-dotcom/truthgate (0 stars, last pushed 1mo ago), licensed MIT. It adds 23 tokens to every session and 334 once invoked, about $0.0001 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

searcharvester-deep-research

Deep-research methodology with a ROLE-BASED team of parallel sub-agents. One delegatetask batch dispatches 2–3 researchers (each on a distinct sub-question), 1 critic (adversarial counter-search), and 1 fact-checker (numeric/date verification). The lead synthesises their findings into a cited report at ./report.md.…

vakovalskii/searcharvester · 107 tokens

searcharvester-extract

Fetch a web page and extract its main content as clean markdown via trafilatura. Strips navigation, ads, and footers automatically. Supports size presets (short/medium/long) and full mode with pagination for very long documents. Use after searcharvester-search to actually read the pages behind your URLs.

vakovalskii/searcharvester · 71 tokens

searcharvester-search

Search the web via SearXNG (100+ engines: Google, DuckDuckGo, Brave, Bing, Yandex, Mojeek, ...) through a Tavily-compatible API. Use when the user needs to find information, recent news, compare sources, gather facts, or locate specific URLs. Returns title + URL + short snippet for each result. For full page content…

vakovalskii/searcharvester · 95 tokens

release

Cut a versioned release and publish everos to PyPI via the tag-triggered workflow.

EverMind-AI/EverOS · 20 tokens

datadesigner-docs

Maintain the NeMo Data Designer Fern docs site under fern/. Use for any documentation change. Triggered by: "edit docs", "add doc page", "update docs", "rename page", "fix broken link", "add redirect", "preview docs", "publish docs", "regenerate notebooks", "update dev note", any request that touches fern/.

NVIDIA-NeMo/DataDesigner · 80 tokens

neuron-structured-output

Design and implement structured output classes for Neuron AI agents using SchemaProperty attributes and validation rules. Use this skill when the user mentions structured output, JSON schema extraction, data validation, output classes, DTOs for AI responses, extracting structured data from LLM, or configuring property…

neuron-core/neuron-ai · 104 tokens