What is dr-validate-change?

A validation step for a completed software change that runs every acceptance check from SPEC.md and records a PASS or FAIL result in VALIDATION.md.

Use it after CHECKLIST.md is complete to rerun the required checks, run the test suite, and record failures without modifying the code.

It provides evidence that the whole change meets its requirements instead of relying only on individual checklist steps.

How AHepi/DeepReason configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to AHepi/DeepReason. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/AHepi/DeepReason/main/.claude/skills/dr-validate-change/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/AHepi/DeepReason

Made for: Claude Code.

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/ahepi/deepreason/dr-validate-change/github.svg)](https://agentmods.dev/skills/ahepi/deepreason/dr-validate-change)

Your own site

<a href="https://agentmods.dev/skills/ahepi/deepreason/dr-validate-change"><img src="https://agentmods.dev/badge/skills/ahepi/deepreason/dr-validate-change/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/ahepi/deepreason/dr-validate-change"><img src="https://agentmods.dev/badge/skills/ahepi/deepreason/dr-validate-change.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 46 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,432 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. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

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	$0.00046	$0.01432
Opus 5.5	$0.00018	$0.00573
Sonnet 5.5	$0.00009	$0.00286
Haiku 4.5	$0.00005	$0.00143

Measured 2mo ago against content hash f8fb987cdf52, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

dr-validate-change 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 2mo 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/skills/dr-validate-change/SKILL.md · 118 lines

How it starts

The opening of the file, as written. The whole thing — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Validate the change

Input: SPEC.md's acceptance checks + the finished checklist. Output: VALIDATION.md with verdict PASS or FAIL. You run checks and record outcomes. You do not fix anything — a failure routes back to re-planning with evidence, which is cheaper than a hidden patch that invalidates the checklist's audit trail.

Procedure

Re-read REQUEST.md, SPEC.md, CHECKLIST.md in full. Yes, again — this is the phase that catches forgotten requirements, and it can only catch what it re-reads.
Run EVERY acceptance check in SPEC.md, in item order, even ones a checklist step already ran — steps prove local progress; this phase proves the assembled whole. Paste each real output.
Run the regression ring: the full gate (pytest tests/ -q -n 4) must end 0 failed. A failure you caused is a FAIL verdict; a pre-existing failure you can prove pre-dates the change (git stash → rerun → git stash pop) is recorded as such and does not block, but goes to PARKED.md.
Behavior-preservation spot-check: if the change touched a reader or validator of the append-only record, re-run verify_root on one known-good committed root and one defect-era root — prior verdicts must be unchanged except where SPEC.md says otherwise.
Frozen-surface diff — paste it, empty or explained:
```
 git diff --stat <tranche-base>..HEAD -- \
   src/deepreason/capabilities/state.py src/deepreason/harness.py \
   src/deepreason/invariants.py src/deepreason/run_manifest.py \
   src/deepreason/qualification.py
```
Empty output is the expected result and is pasted as proof. Non-empty output is a FAIL unless REQUEST.md quotes the operator approving that exact surface — convention guards these files at design time, but this paste is the one MECHANICAL tripwire on the path, so it is not optional.
Packaging-surface check. If the change touched pyproject.toml, CLI entry points, the MCP server surface, or the wheel layout: run python scripts/wheel_smoke.py (plus python -u scripts/wheel_operational_smoke.py when the operational provider-facing surface moved) and paste the last lines. The smokes pin expected sets and hashes; a surface change whose commit did not update those pins is a FAIL. If the surface did not move, write "packaging surface untouched — smoke not owed": the skip must be a recorded decision, not an omission.

Read the full file on GitHub · 118 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.

2mo ago First seen · 118 lines · 46 tokens per session scan A f8fb987cdf52

Subscribe to this mod's changes

dr-validate-change is a skill published in the GitHub repository AHepi/DeepReason (141 stars, last pushed 1mo ago), licensed MIT. It adds 46 tokens to every session and 1,432 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 2026-08-30.

Testing

research-engineer

An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.

davila7/claude-code-templates · 43 tokens

tika-eval-compare

Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".

apache/tika · 50 tokens

cua-driver

Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.

davidondrej/skills · 68 tokens

DeepReason: Skill for Claude Code