Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/arrrrny/zuraffanpx agentmods add commands/arrrrny/zuraffa/speckit.md-doctor.driftWrote 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.
[](https://agentmods.dev/commands/arrrrny/zuraffa/speckit.md-doctor.drift)<a href="https://agentmods.dev/commands/arrrrny/zuraffa/speckit.md-doctor.drift"><img src="https://agentmods.dev/badge/commands/arrrrny/zuraffa/speckit.md-doctor.drift/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.
<a href="https://agentmods.dev/commands/arrrrny/zuraffa/speckit.md-doctor.drift"><img src="https://agentmods.dev/badge/commands/arrrrny/zuraffa/speckit.md-doctor.drift.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00042 | $0.01012 |
| Opus 5 | $0.00021 | $0.00506 |
| Sonnet 5 | $0.00008 | $0.00202 |
| Haiku 4.5 | $0.00004 | $0.00101 |
Grade A, and why
speckit.md-doctor.drift 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 7d 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.
How it starts
The opening of the file, as written. The whole thing — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MD Doctor — Drift (re-evaluate)
The "run it a week later" command. It loads the previous run's facts and ground
truth, computes exactly what changed since then (git commits, changed .md files,
new .memsearch records, refreshed TDD verdicts), and re-grades every previously
scored file so you can say what is newly true, what is newly false, and which past
suggestions were resolved or ignored.
User Input
$ARGUMENTS
Recognized modifiers:
--path <subdir>: re-evaluate only one subtree.--json: emit the delta + re-graded facts as JSON.
With no input, re-evaluate the whole repo against the last run.
Preconditions
init or a prior scan must have run (there must be
.specify/md-doctor/state/last-run.json and facts.json). If missing, stop and tell
the user to run /speckit.md-doctor.init first.
Workflow
Phase 1: Compute the delta
bash .specify/extensions/md-doctor/scripts/bash/md-doctor.sh drift --json > /tmp/md-doctor-delta.json
(Dev fallback: bash <extension-root>/md-doctor/scripts/bash/md-doctor.sh drift --json.)
The delta contains:
prev_run.head/prev_run.timestamp— the last anchored run.commits_since[]— commits between the last run's HEAD and current HEAD: what was implemented since.md_changed_since[]—.mdfiles git changed in that span.memsearch_files_since[]— new daily records: what agents actually did.
Read the delta. Open any memsearch_files_since and skim commits_since subjects —
these are the new ground truth since last week.
Phase 2: Re-score previously graded files
Load .specify/md-doctor/state/facts.json (last run's grades). For each file:
- If it is in
md_changed_since, its content changed → re-extract claims and re-grade from scratch against the new ground truth. - If a
commits_since/memsearch_files_sinceentry resolves a claim the doc was wrong about, upgrade its verdict (e.g.false→staleif the code caught up). - If new ground truth contradicts a doc that was
truthful, downgrade it (true→false/stale). - Mark each file's status relative to last run:
resolved(a pastupdate/deletesuggestion was applied),worsened,improved, orunchanged.
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.
- 7d ago First seen · 98 lines · 42 tokens per session scan A dcc2d57a095a
speckit.md-doctor.drift is a command published in the GitHub repository arrrrny/zuraffa (5 stars, last pushed today), licensed MIT. It adds 42 tokens to every session and 1,012 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 2026-09-03.
Other commands, from other repositories
verify-bug
Post-merge UAT verification workflow. Walks JIRA reproduce steps, performs comparative audits (Before/After), attaches evidence to JIRA, and transitions status on PASS.
codebase-review
Review an entire codebase for architecture, engineering health, and exploitable risk; generate a prioritized remediation plan, an evidence-anchored system knowledge document, or both.
debug
Diagnose Ansible playbook issues from symptoms.
perf-profile.template
This prompt was authored for Claude-style slash workflows. In Codex runtime, adapt tool calls as follows.
building-flutter-apps-skill-ac1294fc
Flutter Riverpod app architecture and Windows installer delivery. Use before changing a Riverpod Flutter app/package or its Windows desktop packaging/update pipeline; skip non-Riverpod stacks and pure-Dart work.
poly-check
Lint and check formatting with poly — apply no fixes; summarize findings and drift.