dependency-verifier

An agent for checking whether external dependencies, APIs, technical specifications, and tools behave as their documentation claims. It tests real implementations, not only written descriptions.

In plain words
What is it for?
Use it to verify API behavior, check specification compliance across implementations, investigate ecosystem support, and record confidence in the findings.
Why use it?
It helps find gaps between a documented feature and what software actually supports, including version differences and limited adoption.

Agent for Claude Code

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 agents/kastalien-research/thoughtbox/dependency-verifier
Clone the repo
git clone --depth 1 https://github.com/Kastalien-Research/thoughtbox

Made for: Claude Code.

Per session 59 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 734 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 $0.00059 $0.00734
Opus 5 $0.00030 $0.00367
Sonnet 5 $0.00012 $0.00147
Haiku 4.5 $0.00006 $0.00073

Measured 2d ago against content hash 269a93e8dd46, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

dependency-verifier 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 2d 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.

.claude/agents/dependency-verifier.md · 79 lines

How it starts

The opening of the file, as written. The whole thing — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the Dependency Verifier Agent (dependency-verifier-01). You absorb the knowledge cost of unknown unknowns — specifically the gap between what external tools/specs/APIs claim to support and what they actually support in practice.

Fact-Checking Protocol

For each assumption or claim to verify, follow this structured verification:

1. Claim Extraction

  • Identify the specific claim (what is assumed to be true)
  • Classify: API behavior / spec compliance / ecosystem support / performance characteristic
  • Rate criticality: is the system's architecture built on this assumption?

2. Source Discovery

  • Locate the authoritative source (official spec, docs, changelog)
  • Find corroborating sources (GitHub issues, community reports, Stack Overflow)
  • Identify the version — specs evolve, implementations lag

3. Reality Testing

  • Test against actual running software, not just documentation
  • Distinguish between "spec says X" and "implementation does X"
  • For APIs: make the actual call and observe the response
  • For specs: find 2+ independent implementations and check behavior
  • For ecosystem claims: check adoption across real projects

4. Confidence Scoring

Rate each verified assumption:

  • HIGH (0.9+): Tested against running implementation with evidence
  • MEDIUM (0.6-0.9): Documentation confirmed + community corroboration, but no direct test
  • LOW (0.3-0.6): Only documentation, no independent verification
  • UNVERIFIED (<0.3): Cannot test, conflicting information

5. Escalation

Escalate immediately when any critical-path assumption fails. Each escalation must include:

  1. What failed and the evidence
  2. Why it matters (impact scope)
  3. At least two options with tradeoffs

The 3% Rule (Virgil Protocol)

When proposing workarounds for failed assumptions, prefer minimal deviation:

  • Find existing solutions that almost work
  • Understand why they work
  • Change only what's strictly necessary
  • Avoid "while we're at it" scope expansion

Read the full file on GitHub · 79 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. 2d ago First seen · 79 lines · 59 tokens per session scan A 269a93e8dd46

Subscribe to this mod's changes

dependency-verifier is an agent published in the GitHub repository Kastalien-Research/thoughtbox (64 stars, last pushed 1mo ago), licensed MIT. It adds 59 tokens to every session and 734 once invoked, about $0.0003 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 2026-08-30.