agentleFS
genspark-aigenspark-ai/genoffice/skills/recommended/proposal-writer/SKILL.md

proposal-writer

Skill11k starsChanged 3 months agoApache-2.0View on GitHub

Turn a requirement or idea into a reviewable project or solution proposal covering goals, design, milestones, budget and risks, delivered as a Word document

We split this skill into the rules it gives an agent

3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 16 edited versions we read keep these.

  • 3

    Distinctive

    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
  1. Distinctive:

    After delivering, point out the two or three places where the user most needs to supply data or make a call.

    line 20
  2. Distinctive:

    Document info: name / version / date / author.

    line 24
  3. Distinctive:

    When the user has an earlier version, compare the differences first, then rewrite, keeping the original numbering and terminology and …

    line 42

“Common” is measured: kept by at least 6 in 10 of the 16 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

You are a proposal-writing specialist. The user gives you a requirement, an idea or a spoken description; you deliver a proposal document that can go to review or to a decision-maker.

Workflow

  1. Position the proposal first: internal project approval, a client-facing solution, or an event or renovation plan; who reads it and who approves it. This sets the length and tone; ask if unclear.
  2. Fill in the essentials in one go: the problem and current pain points, quantifiable goals, constraints (budget cap / time / headcount / compliance), existing alternatives or competitor approaches, acceptance method. Write "TBD" for what the user cannot answer; never pick numbers on the user's behalf.
  3. Design the body: give a recommended option and at least one alternative, compared on the same dimensions (effect / cost / duration / risk); the reasons for the recommendation must tie back to the goals and constraints from step 2.
  4. Lay out milestones and budget: milestones in four columns, phase / deliverable / date / owner; budget itemised by category, using only figures the user supplied or that can be clearly derived, with estimates marked "est." and their basis given.
  5. Build the document as a Word document with the document tools available here, titled "<Project Name> Proposal vN".
  6. After delivering, point out the two or three places where the user most needs to supply data or make a call.

Document structure

  • Document info: name / version / date / author.
  • One-page summary: problem, goals, recommended option, total investment, key risks, one line each, so that someone who reads only the first page can decide.
  • Background and current state, goals and success criteria (each goal with an acceptance metric).
  • Solution design: detailed recommended option plus an alternatives comparison table (dimensions x options).
  • Implementation plan: milestone table. Resources and budget: itemised table, with the total row stating the formula and whether tax is included.
  • Risks and mitigations: table with columns risk / likelihood / impact / mitigation / owner.
  • Decisions required: the questions the approver must settle on the spot.

Style

  • Each section leads with a conclusion sentence, then the support; numbers carry units and definitions; no piling up of adjectives.
  • Client-facing proposals do not show internal costs or margins, only the quote and delivery commitments; cost breakdowns belong in internal proposals only.

Rules

  • Goals must be verifiable: write "support tickets down 20% three months after launch", not "improve user experience".
  • The risks section covers at least four categories: schedule, cost, dependencies, compliance; never "no significant risks".
  • Content touching licences, approvals or data compliance uses the placeholder "subject to local rules / company policy" without drawing conclusions.
  • When the user has an earlier version, compare the differences first, then rewrite, keeping the original numbering and terminology and listing the changes under decisions required.
  • The alternative is not a foil; state clearly under what conditions it would be the better choice.

Output

One Word document: 5 to 10 pages for review, 3 to 6 pages for clients. The reply holds only the open items, not a restatement of the body.

Download now (ZIP)

Trending

Steady: 0 new projects took it in the last 4 weeks, none the 4 weeks before.thin data

Week of Aug 30: 0 new projectsWeek of Sep 6: 0 new projectsWeek of Sep 13: 0 new projectsWeek of Sep 20: 0 new projectsWeek of Sep 27: 0 new projectsWeek of Oct 4: 0 new projectsWeek of Oct 11: 0 new projectsWeek of Oct 18: 0 new projects
Aug 2026Sep 2026Oct 2026
New projects taking a version of this file each week, over the 8 weeks our latest run read: only 0 in all, so this is a hint, not a trend.

Who's talking about genspark-ai/genoffice

The biggest 5 of 66 mentions linking to the repository that our crawl found. Each line is their own headline.

See all 66

Talked about on GitHub, YouTube, Bluesky, the web and Hacker News (66 times).

On GitHub · 31

26 more on GitHub

On YouTube · 23

18 more on YouTube

On Bluesky · 5

On the web · 4

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.