workflow-audit SKILL.md skillB
workflow-audit from abhiramdwivedi/claude-ops is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add abhiramdwivedi/claude-ops/workflow-audit@git:20261008.661d3a6Pins git:20261008.661d3a6 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- B, 16 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- abhiramdwivedi/claude-ops
- Path
plugins/workflow-audit/skills/workflow-audit/SKILL.md- Versions recorded
- 5
- Latest version
- git:20261008.661d3a6, committed Oct 8, 2026
- First version recorded
- Oct 6, 2026
- Size
- 16,095 bytes
- Repository stars
- 3
- License
- MIT
Audits how the person uses Claude Code from their own local transcripts and writes a blunt, evidence-graded report of the few changes that would matter most, each with one measurable experiment. Use when the person runs /workflow-audit, or asks to audit, review or get honest feedback on how they use Claude Code, whether they are using it well, or where their tokens or rework go.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Workflow audit You are auditing how this person works with Claude Code. Your job is to tell them what is true about their own usage, backed by their data, not to reassure them. People rate their own AI use far higher than measurement does: in METR's 2025 trial, developers believed they were 20% faster and were measured 19% slower. So nothing in this report comes from what the person believes about themselves. It comes from their transcripts. A script does the measuring. You do the judging. The catalog says what current advice is, who gave it, when, and how strong the evidence is. ## Ground rules - **Evidence or silence.** Every claim cites a number from `metrics.json` (with its n) or a sample excerpt. If the data can't support a claim, don't make it. - **No score, no praise section, no warm-up.** A "keep doing" line is allowed only when dropping the habit would cost something you can point to. - **Their own history is the baseline.** "Your sessions with X had 3x the corrections" beats "best practice says X". Comparisons need enough sessions on both sides; show the counts. …
Read the whole file at its exact version.
How to install
mdr add abhiramdwivedi/claude-ops/workflow-audit@sha256:16584449be493a59The line at the top of this page pins the label git:20261008.661d3a6, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_kvkso7nqxj2b7gsw)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 5 versions of this file, between Oct 6, 2026 and Oct 8, 2026, more than one a day.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261008.661d3a6 latest | 2026-10-08 | 661d3a6 | 16,095 B | B | view · diff |
| git:20261006.235c4e8 | 2026-10-06 | 235c4e8 | 13,886 B | B | view · diff |
| git:20261006.5ab5a29 | 2026-10-06 | 5ab5a29 | 13,183 B | B | view · diff |
| git:20261006.68023ac | 2026-10-06 | 68023ac | 11,644 B | B | view · diff |
| git:20261006.d506e56 | 2026-10-06 | d506e56 | 9,129 B | B | view |
Audit of the latest version
- fail: No script tag (matched: <script)
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (16095 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
Starting a new one? Use a template for SKILL.md on mdpantry, which passes every one of these checks and says what to change in it.
Source
abhiramdwivedi/claude-ops · 3 stars · license MIT · pushed 2026-11-02 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_kvkso7nqxj2b7gsw GET https://markdownregistry.com/api/v1/resolve?ref=abhiramdwivedi/claude-ops/workflow-audit GET https://markdownregistry.com/api/v1/blob/16584449be493a59a780994f6ad09530eb108ed984d8b6739f5cc49e7a32f527
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.