deliverable-qa
Skill4 starsChanged 36 days agoMITView on GitHub
QA a client deliverable, report, or presentation against consulting quality standards before it goes out. Use when asked to review, QA, proofread, or sanity-check anything client-facing. Usage - /deliverable-qa path/to/deliverable
In this repository11 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- change-requestSkill
- code-review-lightSkill
- exec-summarySkill
- meeting-prepSkill
- sow-reviewSkill
- stakeholder-prepSkill
- status-commsSkill
- status-reportSkill
- transcript-actionsSkill
- weekly-briefSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 47
After reading this, does the client know what to do next?
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Deliverable QA
Review the specified deliverable against consulting quality standards and produce a structured QA assessment. This is an internal review meant to improve quality before the deliverable reaches the client. It is a second reviewer, never the last one.
Input
$ARGUMENTS is the path to the deliverable (.md, .docx, .pptx, .pdf, or .xlsx). If no
path was given and the deliverable was drafted earlier in this conversation, say so and offer to
review it anyway, but note that a review of your own draft is weaker than a cold read. For a
genuinely fresh read, the user can save the draft to a file and either run /clear and this skill
again, or ask for the second-reviewer subagent.
QA Dimensions
Assess the deliverable across these five dimensions:
1. Executive Clarity
- Does the document lead with the "so what"? Would a C-suite reader understand the key message in 30 seconds?
- Are recommendations concrete and actionable, not just observations?
- Is the language free of jargon that requires insider knowledge?
2. Logical Structure
- Does the argument flow? Does each section connect to the one before and after?
- Are conclusions supported by evidence presented in the document?
- Is there a clear narrative thread from problem to finding to recommendation?
3. Accuracy & Completeness
- Are any key sections missing (for example, no risk section, no next steps)?
- Are data points cited or attributed? Flag any that appear unverified.
- Are there internal inconsistencies (a number in a table that does not match the text, a RAG color that does not match the narrative)?
4. Client-Readiness
- Is the formatting professional and consistent?
- Are there any placeholders, TODOs,
[NEEDS INPUT]markers, or draft language that was not removed? - Does it name any other client, or carry a sensitive detail that should not travel?
- Is the tone appropriate for the stated audience?
5. Actionability
- After reading this, does the client know what to do next?
- Are next steps specific, with named owners and dates?
- Would a client who has not been in the room understand what is being asked of them?
Output Format
Overall Readiness: [✅ Ready / ⚠️ Needs Revision / 🚫 Not Ready]
Strengths: 2 to 3 things the document does well
Issues Requiring Revision:
| # | Issue | Dimension | Severity | Recommendation |
Severity: High (blocks delivery) / Medium (should fix before sending) / Low (nice to have)
Quick Wins:
- Up to 3 small edits that would noticeably improve quality with minimal effort
Outside my ability to verify:
- Anything the review could not check, such as whether a quoted figure is current or whether a technical claim is correct. The delivery lead routes these to someone who can.
Be specific and constructive. Point to the page, slide, or paragraph. Do not rewrite the deliverable unless asked; the owner decides what changes.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

