- Home
- Skills
- Writing & Content
- PM Doc Slop Check
Works with the AI Agents you already use
PM Doc Slop Check
An anti-slop pass built for product docs. Get a score, every vague or ownerless line with the reason, the missing PRD elements and a rewrite that keeps your meaning.
$5
PM Doc Slop Check
Example session with this skill installed
"Slop check this" plus a monthly update that opens with "In today's fast-paced learning landscape", reports "Status: Green", "It was decided to push the mobile launch" and ends with "Any support from leadership would be appreciated."
- Read your context and instructions
- Compiled the pm doc slop
Score 10/100 (Rewrite). 7 high flags, including the soft ask, the passive decision with no date or direction, and a Green status with a delayed launch. 4 missing elements: headline, what changed, decisions, next milestone. A rewrite with a headline, real decision lines and a concrete ask, plus a list of 13 placeholders the author has to fill. A second example shows a solid PRD scoring 87 with three small fixes. Both full reports are included.
Connects securely to your tools. The creator never sees your data.
About this skill
General humanizers make text sound less robotic. Product docs have a different problem: they sound busy and still leave out the number, the owner, the decision or the ask. PM Doc Slop Check reads a PRD, stakeholder update, strategy memo or slide text the way a demanding reviewer would. It checks every sentence against 15 patterns, from throat-clearing openers and hedge stacks to passive ownership ("it was decided"), soft asks and status with no evidence. Then it checks the doc as a whole: does the PRD have a problem with evidence, a metric with baseline, target and date, scope out, risk owners and an ask? You get a 0 to 100 score with the arithmetic shown, a table of flagged lines with exact quotes and fixes, and a rewrite. The rewrite never invents facts. Where a number or owner is missing, it leaves an [ADD: ...] placeholder for you to fill.
What it does
- Detects 15 product-doc patterns: filler openers, vague claims, hedge stacks, buzzword stacking, fake structure, missing numbers, missing decisions, passive ownership, generic AI tells, solution posing as the problem, unmeasurable metrics, scope blur, risks without a plan, soft asks, status without direction
- Checks required elements per doc type: PRD, stakeholder update, strategy doc, deck text
- Scores 0 to 100 with a transparent breakdown and four bands (Ready, Light edit, Needs work, Rewrite)
- Lists flagged lines with unit ID, exact quote, pattern, severity, reason and fix
- Rewrites the doc in your voice and structure, with placeholders where facts are missing
- Lists lines it kept on purpose, such as honest, explained uncertainty
- Modes: full, flags only, rewrite only, strict
Why this beats prompting it yourself
A plain "make this better" prompt polishes the words and quietly fills gaps with plausible numbers. This skill separates wording problems from missing facts, never invents a fact, and shows its scoring so you can argue with it.
Who it's for
- Product managers sending PRDs and updates to busy leaders
- Founders writing investor or board updates
- Product leaders and coaches reviewing their team's docs
- Anyone who drafts with AI and wants the result to sound like a person who knows the product
Use cases
- Clean up a monthly stakeholder update before it goes out
- Check a PRD before a design or engineering review
- Turn a wordy strategy memo into one that states the choice
- Coach a junior PM with concrete, line-by-line feedback
What you get
SKILL.mdwith the 9-step workflow and rewrite rules- A pattern library: 15 patterns with cues, severity and 30 before/after pairs
- Doc type checks for PRDs, updates, strategy docs and decks
- Scoring rules and a report template
- Two complete sample runs
Frameworks and tools
Works in any agent that reads SKILL.md: Claude Code, Cursor, Codex and similar. No API keys, installs, scripts or network access.
Permissions
- Read: the doc you paste or point to
- Write: nothing by default. It returns the report in chat
- Network: none
Known limitations
It checks writing and completeness, not whether the product idea is good. It cannot verify numbers you give it. Very long docs (over 4,000 words) are checked section by section, starting with goals, scope, risks and asks.
It checks writing and completeness, not whether the product idea is good. It cannot verify numbers you give it. Very long docs (over 4,000 words) are checked section by section, starting with goals, scope, risks and asks.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 21 days ago