DeepReason: Skill for Claude Code

.claude/skills/dr-deliver-change/SKILL.md

dr-deliver-change is a skill for Claude Code from AHepi/DeepReason. It costs 48 tokens per session (1,065 once invoked), scanned A, original, MIT.

A delivery process for finishing a validated code change. It checks that the final commit is pushed, compares the result with every original requirement, and writes a delivery report.

In plain words
What is it for?
Use it to close a change tranche, verify the branch and working tree, reconcile requirements against the implementation, record assumptions and parked items, and create DELIVERY.md.
Why use it?
It makes missing, assumed, or deferred work visible before a change is handed over. It only runs after validation has passed.

Skill for Claude Code

Written for Claude Code: installed under .claude/. Also seen: mentions CLAUDE.md.

This is AHepi/DeepReason's own configuration. It tells Claude Code how to work on DeepReason itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything DeepReason configures →

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-deliver-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.

agentmods badge for dr-deliver-change

README.md
[![agentmods](https://agentmods.dev/badge/skills/ahepi/deepreason/dr-deliver-change.svg)](https://agentmods.dev/skills/ahepi/deepreason/dr-deliver-change)
Your own site
<a href="https://agentmods.dev/skills/ahepi/deepreason/dr-deliver-change"><img src="https://agentmods.dev/badge/skills/ahepi/deepreason/dr-deliver-change.svg" alt="Measured on agentmods" height="20"></a>
Per session 48 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,065 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00048 $0.01065
Opus 5 $0.00024 $0.00532
Sonnet 5 $0.00010 $0.00213
Haiku 4.5 $0.00005 $0.00106

Measured 8d ago against content hash 7d9492b01a6b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

dr-deliver-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 8d 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-deliver-change/SKILL.md · 91 lines

How it starts

The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Deliver the change

Input: a PASS VALIDATION.md. Output: everything pushed, and DELIVERY.md — the report the operator actually reads. Delivery is reconciliation, not celebration: its job is to make any gap between what was asked and what was done impossible to miss.

Procedure

  1. Final tree check: git status --porcelain must be empty after a last commit+push of the tranche directory; confirm the branch head exists on origin (git rev-parse HEAD origin/<branch> — one hash).
  2. Build the reconciliation table from REQUEST.md's verbatim requirements — walk EVERY R number, including amendments and superseded ones (superseded rows say so). Allowed dispositions:
    • done — with the commit hash and the acceptance output pointer
    • done-with-assumption — cite A; the operator may override
    • deferred — ONLY with the operator's quoted words permitting it
    • not-done — forbidden here; that is a FAIL, go back
  3. Surface the assumptions (from VALIDATION.md) and PARKED.md contents as explicit lists. Parked items are offered as candidate next tranches, never silently promised — each with its ready-to-send prompt (the park-with-prompt rule), and the close RECOMMENDS one next item from the queue, so the operator answers go/no-go rather than authoring the follow-up.
  4. Report the map delta. One short section: which docs/map/ documents this tranche changed or created, how many checks it added, and any document --stale still lists with the reason it was left. The operator needs to know whether the next reader of this subsystem will be told the truth. "No map change" is a legitimate answer for a tranche that changed no behaviour — say it rather than omitting the section, so its absence is never ambiguous.
  5. Errata check — mandatory, before DELIVERY.md is committed. Did this tranche find any committed document's claim (a handover, a map document, a RESULTS.md, a spec, CLAUDE.md — anything docs/ERRATA.md covers) to be wrong? If yes, the docs/ERRATA.md entry lands in the SAME commit as DELIVERY.md. If no, state "errata: none" explicitly in DELIVERY.md's Errata section — state it, do not omit the section. Same state-not-silence pattern as step 4's map delta: an absent section is ambiguous, an explicit "none" is not.
  6. Write DELIVERY.md leading with the outcome in plain sentences a reader who saw none of the work can follow: what changed, where, how it is proven. No process narration ("first I read the file...").
  7. If the request touched experiments/live evidence: append the dated segment to the relevant RESULTS.md per that record's honest-ledger style ("accepted does not mean true"; never claim more than the record shows). Commit and push it.

Read the full file on GitHub · 91 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. 8d ago First seen · 91 lines · 48 tokens per session scan A 7d9492b01a6b

Subscribe to this mod's changes

dr-deliver-change is a skill published in the GitHub repository AHepi/DeepReason (142 stars, last pushed yesterday), licensed MIT. It adds 48 tokens to every session and 1,065 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 2026-08-30.