prompt-pre-editing SKILL.md skillA
prompt-pre-editing from jgwill/miadi-orchestration-kit is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add jgwill/miadi-orchestration-kit/prompt-pre-editing@v0.4.0Pins v0.4.0 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- jgwill/miadi-orchestration-kit
- Path
claude/miadi-witness/skills/prompt-pre-editing/SKILL.md- Versions recorded
- 5
- Latest version
- v0.4.0, committed Oct 6, 2026
- First version recorded
- Oct 6, 2026
- Size
- 10,935 bytes
- Repository stars
- 5
- License
- none declared
Pre-edit a person's draft prompt for another agent session. Find what will make the run fail (a thing named but never described, a phrase with several readings, unranked goals, sources with no stated role, asks that return what is already known, an act with no owner, no output contract) and hand back the whole rewritten prompt ready to paste. After it is sent, follow the run, judge the result against expected results committed beforehand, and change this skill or the skill the prompt invoked…
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
# Prompt pre-editing **The name.** In machine translation, *pre-editing* is a person rewriting a source text before a machine processes it. Typical rewrites make references explicit, remove ambiguity and give one instruction per sentence, which is what controlled languages prescribe. Here the agent is the machine and the person's draft is the source text. Offering the rewrite back so the person can send it or correct it is what conversation analysis calls a *formulation*: saying back the gist and upshot of what someone said so they can confirm it (Heritage and Watson, 1979). Earned 2026-10-06. The worked example is `examples/261006-factory-vocabulary.md`, with the draft, the defects, the rewrite, the expected results and the result. ## 1. Read before judging - Read the draft in full. An unsent draft sits in the other session's composer: `tmux capture-pane -p -J -t <session> -S -200`. A long composer shows only its last lines. Say which part you could not see. Never type into that pane (`dispatch-discipline` §4). - Read everything the draft points to, such as other panes, files, reviews and packets. The missing referent is usually there. …
Read the whole file at its exact version.
How to install
mdr add jgwill/miadi-orchestration-kit/prompt-pre-editing@sha256:bae956c97f9ea922The line at the top of this page pins the label v0.4.0, 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_7tazlwf5tuyuaeco)
1 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 5 versions of this file, all committed on Oct 6, 2026.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| v0.4.0 latest | 2026-10-06 | 23a2d4a | 10,935 B | A | view · diff |
| v0.3.1 | 2026-10-06 | 67782fb | 10,341 B | A | view · diff |
| v0.3.0 | 2026-10-06 | f036aaa | 9,767 B | A | view · diff |
| v0.2.0 | 2026-10-06 | efd06fb | 8,679 B | A | view · diff |
| v0.1.0 | 2026-10-06 | 008d0cf | 6,728 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (10935 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
- pass: No script tag
Source
jgwill/miadi-orchestration-kit · 5 stars · license none declared · pushed 2026-10-21 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_7tazlwf5tuyuaeco GET https://markdownregistry.com/api/v1/resolve?ref=jgwill/miadi-orchestration-kit/prompt-pre-editing GET https://markdownregistry.com/api/v1/blob/bae956c97f9ea92256c54c92e76620798f5afe06f8ac0e237d07d47ef47b112d
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.