Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    SlideOS & Decision Brief Pack

    1

    This pack measures those faults so an agent can fix them and show the exit code.

    $24

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    SlideOS & Decision Brief Pack

    SlideOS & Decision Brief Pack

    Example session with this skill installed

    Lint bad.md slides for dense bullets and unsourced metrics, verify bad.html decision brief portability and print rules, and check bad.srt subtitle reading speed.

    • Read your context and instructions
    • Compiled the slideos & decision

    Real output from running the packaged scripts on small fixtures.

    1. node slide_deck_linter.mjs bad.md

    $ node slide_deck_linter.mjs bad.md
    bad.md: 4 slide(s), 7 finding(s)
      [warn] slide 1 too-much-text - 313 characters of body text (limit 220).
          fix: Cut to the one sentence the audience must remember.
      [warn] slide 1 empty-heading - Heading "Overview" says nothing a reader can disagree with.
          fix: State the takeaway: what changed, by how much, so what.
      [warn] slide 1 missing-source - Numeric claim without a source, date or sample size.
          fix: Add "Source: name, year" or the measurement (n, period) on the slide.
      [error] slide 2 empty-heading - Slide has no heading.
          fix: Write the heading as the slide's claim in a full phrase.
      [warn] slide 3 uniform-card-repeat - 4 near-identical cards/bullets (length spread 0.03).
          fix: Make the one that matters bigger, or turn the set into a table, chart or sequence.
      [warn] slide 4 missing-source - Numeric claim without a source, date or sample size.
          fix: Add "Source: name, year" or the measurement (n, period) on the slide.
      [warn] slide 4 multiple-claims - Heading joins two ideas and the body argues both.
          fix: One slide, one claim. Split it.
    (exit 1)
    

    2. node decision_html_checker.mjs bad.html

    $ node decision_html_checker.mjs bad.html
    bad.html: 10 finding(s)
      [error] line 6 conclusion-first - No conclusion or decision request before the first table.
          fix: Put one sentence first: the decision you ask for and the reason.
      [error] line 1 problem-issue-action-table - No 3-column table with problem, issue and action columns.
          fix: Add a table: Problem | Issue (why it happens) | Action (what to do next).
      [error] line 2 external-cdn - <link> loads a remote resource: //cdn.example.test/a.css
          fix: Inline it or vendor it; a brief must open offline.
      [error] line 1 table-narrow-560 - Tables have no rule for screens up to 560px wide.
          fix: Add @media (max-width: 560px) { table, tr, td, th { display: block; } } or wrap the table in an overflow-x: auto box.
      [warn] line 1 table-overflow-wrap - No overflow-wrap rule: long words or URLs can push the table wider than the screen.
          fix: Add overflow-wrap: anywhere to td, th.
      [error] line 5 print-table-override - Print rule turns table parts into blocks (table): the printed table loses its columns.
          fix: Do not copy the mobile table override into @media print; keep display: table / table-row / table-cell.
      [warn] line 5 print-border-left - border-left is redefined in print (.card): the accent bar shifts layout or disappears.
          fix: Keep the same border width as screen and only change colour; add print-color-adjust: exact.
      [warn] print-color-adjust-missing - Coloured borders or fills are used but print-color-adjust: exact is not set.
          fix: Add -webkit-print-color-adjust: exact; print-color-adjust: exact; to body.
      [error] line 6 body-background-missing - body has no explicit background colour (a background on html alone is not enough).
          fix: Set background: #ffffff (or your colour) on body, or on html, body together.
      [warn] line 1 viewport-missing - No viewport meta tag.
          fix: Add <meta name="viewport" content="width=device-width, initial-scale=1">.
    (exit 1)
    

    3. node srt_pacing_auditor.mjs bad.srt

    $ node srt_pacing_auditor.mjs bad.srt
    bad.srt: 3 cue(s), average 10.5 chars/sec, 7 finding(s)
      [warn] cue 1 (line 2) cps-too-high - 17.5 chars/sec (limit 17).
          fix: Trim a few words or hold the cue longer.
      [error] cue 2 (line 6) cps-too-high - 88.6 chars/sec (limit 17).
          fix: Shorten the text or extend the cue; viewers cannot read it in time.
      [warn] cue 2 (line 6) duration-too-short - Shown for 0.70s (minimum 1s).
          fix: Merge with a neighbour or extend the end time.
      [warn] cue 2 (line 6) line-too-long - Line is 62 columns wide (maximum 42): "Overlap and too fast for any h"
          fix: Break the line at a phrase boundary.
      [error] cue 2 (line 6) time-overlap - Overlaps the previous cue by 0.500s.
          fix: Move the start after the previous end or shorten the previous cue.
      [error] cue 3 (line 10) time-reversed - End time 11.000s is not after start 12.000s.
          fix: Fix the time codes.
      [warn] cue 3 (line 10) silence-gap - 8.8s with no caption before this cue.
          fix: Confirm the gap is intended silence, or add the missing line.
    (exit 1)
    

    Clean runs (exit 0)

    $ node slide_deck_linter.mjs clean.md
    clean.md: 2 slide(s), 0 finding(s)
    (exit 0)
    
    1. node slide_deck_linter.mjs bad.md

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Stop taking the agent's word for it

    Decks, one-page briefs and captions fail in repeatable ways: walls of bullets, identical cards, "Overview" headings, numbers with no source, tables that break on a phone or in print, subtitles that flash by unreadably. This pack measures those faults so an agent can fix them and show the exit code.

    Each skill ships one self-contained Node.js script (Node 18+, standard library only). It prints findings or results with a fix, and exits non-zero when the gate fails. Add --json for machine-readable output.

    The 3 skills

    1. anti-slop-slide-deck-architect plans and lints a slide deck (Markdown, JSON or HTML):

    • one-claim-per-slide method
    • too much text and too many bullets
    • near-identical card grids
    • empty headings, numbers with no source, several claims on one slide

    2. executive-one-page-html-brief checks a one-page HTML decision brief:

    • conclusion before the first table, and a Problem / Issue / Action table
    • zero remote dependencies and a 560px narrow-screen table rule
    • print traps: table display overrides and border-left changes in @media print
    • explicit body background; includes a pitfalls reference

    3. media-pacing-and-caption-gate audits SRT and VTT captions:

    • characters per second above the limit
    • cues that are too short or too long
    • overlapping or reversed time codes and long silent gaps
    • line width (full-width characters count double) and line count

    Built for agents and for review

    • Single file per skill, standard library only: copy one script anywhere and run it
    • Works when the skill is loaded through the Agensi MCP: write the script to a temporary file, run it, delete it
    • No network access, no shell commands, no environment reads
    • Exit code and counts you can quote as evidence

    What is in the ZIP

    • README.md
    • shared/verification-principles.md
    • skills/anti-slop-slide-deck-architect/scripts/slide_deck_linter.mjs
    • skills/anti-slop-slide-deck-architect/SKILL.md
    • skills/executive-one-page-html-brief/references/print-and-mobile-pitfalls.md
    • skills/executive-one-page-html-brief/scripts/decision_html_checker.mjs
    • skills/executive-one-page-html-brief/SKILL.md
    • skills/media-pacing-and-caption-gate/scripts/srt_pacing_auditor.mjs
    • skills/media-pacing-and-caption-gate/SKILL.md

    Character counts are a proxy for density; a dense, well-structured slide can be fine.
    The HTML checker is static: it reads the file but does not render it. Check the page at 560px and in print preview too.
    The caption auditor measures timing and text length, not whether the words match the audio.

    How to use

    1. 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. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 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 today

    Requires

    Node.js 18+
    Node.js

    Permissions

    Read Files
    Terminal / Shell
    Listedtoday

    What's inside

    Frequently Asked Questions

    No. The scripts only read local files and print to the terminal. They contain no network, shell, or environment-variable code.

    Only Node.js 18 or newer. There is no package install.

    Yes. Each script is a single file with no imports of its siblings.

    Yes. Each SKILL.md explains the temporary-file procedure: write the script out, run it, delete it.

    New in Writing & Content