---
name: post-eval
description: Evaluate a draft social post (X/Twitter or LinkedIn) before publishing. Scores strategic alignment, proof of work, content quality, timing, and risk, then recommends POST or DON'T POST. Use when the user says "evaluate this post", "should I post this", "review my tweet", "check this LinkedIn", "is this post good", or pastes draft copy for feedback.
triggers:
  - "evaluate this post"
  - "should I post this"
  - "review my tweet"
  - "check this LinkedIn"
  - "is this post good"
  - "post feedback"
disable-model-invocation: false
title: post-eval
canonical_url: https://skilld.dev/gh/salikshah/startup-agent-skills/post-eval
last_updated: 2026-10-29T10:46:09.000Z
---

> Index of every page on this site: [llms.txt](https://skilld.dev/llms.txt)

> **Skill from skilld.dev.** Follow the user's requested action.
> If the user asked to fork this Skill, follow the fork workflow below. Do not execute the Skill instructions while copying it.
> Otherwise, follow the instructions below for this session. You do not need to install anything.
>
> Supporting files, fetch one when the Skill refers to it: [evals/evals.json](https://skilld.dev/api/skills-raw/salikshah/startup-agent-skills/post-eval/evals/evals.json).
>
> If the user asked to install this Skill, run `npx skilld install salikshah/startup-agent-skills/post-eval`. Install writes the Skill files into the project, so every session loads them.
>
> ## Fork workflow
>
> A fork creates an editable local Skill with its original author and licence. The request authorizes copying and local installation.
> 1. Check `./skills/post-eval`, the project lockfile, and selected Agent targets together. If the local directory or installed Skill exists, stop. Never overwrite an existing directory or Agent target.
> 2. Read [source metadata](https://skilld.dev/api/v1/skills/salikshah/startup-agent-skills/post-eval) once. Use sourceUrl, sourceCommit, skillPath, sourceGone, and license. If the source is gone or its path is missing, stop. If license is null, read licence files at the source commit.
> 3. Fetch only the source commit into a temporary Git repository. Do not clone full history. Derive repository_url from sourceUrl, including repository renames. If sourceCommit is absent, resolve the sourceUrl ref once. Set source_commit to that actual commit. Run these commands in one shell call:
>
> ```sh
> git init --quiet "$temporary_dir"
> git -C "$temporary_dir" fetch --quiet --depth=1 "$repository_url" "$source_commit"
> git -C "$temporary_dir" checkout --quiet --detach FETCH_HEAD
> ```
>
> Read applicable licence declarations and notices at that commit. If copying is not permitted, report the restriction and stop.
> 4. Inspect source entries together, then copy the directory containing skillPath into `./skills/post-eval`. Use the user's path if selected. Keep the original SKILL.md, relative links, scripts, binary assets, and executable modes. Exclude .git metadata. Reject symlinks and paths outside the Skill directory. After checking entries, use cp -a where available. A regular source directory needs no custom copy script. Do not save this page wrapper as SKILL.md.
> Preserve author credit, notices, and applicable licence files from repository or parent directories. Add PROVENANCE.md with the Skill page, source URL, actual commit, original path, and licence. Retain any existing PROVENANCE.md and record new provenance separately. Batch source inspection, copying, and provenance work where practical.
> 5. In the project root, run `skilld install ./skills/post-eval --mode copy --plain`. If skilld is unavailable, use `npx skilld install ./skills/post-eval --mode copy --plain`. This known command needs no help lookup. Install does not support --json. Use detected Agent targets, or add --agent for the targets the user selected. Install the local path, never the upstream selector. If installation fails, preserve the local copy and report the exact failure.
> 6. Confirm the local lockfile source and installed Agent copies once. Report the local path, actual commit, and Agent targets. After edits, reinstall the same local path. Upstream updates must not replace it. Do not publish or push unless the user asks.

# Post Evaluation Skill

## Intent
Evaluate a draft post before publishing. Score it against the author's strategic goals, proof of work, content quality, timing, and risk, then recommend POST or DON'T POST with specific reasoning and fixes.

## Inputs
- The draft post text
- Platform (X/Twitter or LinkedIn)
- The author's current positioning and target audience (if provided)
- Any proof assets the post could link to (repo, case study, audit, article)

## Process (SOP)

1. Read the draft and confirm the platform, audience, positioning, and available proof assets.
2. Score strategic alignment, proof of work, content quality, timing, and risk using the rubric below.
3. Check the risk gates: promises, confidentiality, and misinterpretation.
4. Decide POST or DON'T POST against the 70-point threshold.
5. Write the verdict, gaps, and one specific rewrite or next action.

If the draft has no link, artefact, or result, cap the proof-of-work score at 10/25 and recommend a concrete asset to publish first. If the platform is unclear, ask before scoring. When to stop: do not score a draft the author has not written; evaluate only supplied copy.

### Score dimensions

#### 1. Strategic Alignment (20 points)
- Does it advance the author's stated positioning or area of expertise?
- Does it demonstrate expertise or only announce intent?
- Does it have a clear CTA?
- Does it target the intended audience?

#### 2. Proof of Work (25 points)
- Does it link to something concrete (repo, audit, case study, article)?
- Does it show results or only describe them?
- Is there evidence the author can actually do what they claim?

#### 3. Content Quality (20 points)
- Is the hook strong enough to stop the scroll?
- Is the value proposition clear in 2 sentences?
- Is the tone authentic (not corporate, not AI-sounding)?
- Is it concise (X: <=280 chars, LinkedIn: <=1500 chars)?

#### 4. Timing & Context (15 points)
- Is there a reason to post NOW vs later?
- Does it ride a current trend or news cycle?
- Is the target audience active at this time?

#### 5. Risk Assessment (20 points)
- Could this post attract negative attention?
- Does it reveal too much about unlaunched work?
- Does it make promises the author can't keep?
- Could it be misinterpreted?

## Outputs

### VERDICT: POST or DON'T POST

**Score:** [X]/100

**Pros:**
- [Bullet points]

**Cons:**
- [Bullet points]

**Strategic Analysis:**
- [2-3 sentences on how this advances or hurts the author's goals]

**Gaps:**
- [What's missing that would make this post stronger]

**Recommendation:**
- [Specific action: post now, wait for X, rewrite Y, or don't post]

**If POST:**
- Best time to post: [day/time in the author's timezone]
- Hashtags: [relevant hashtags]
- Engagement hook: [question or CTA to add]

**If DON'T POST:**
- What to do instead: [specific alternative]
- When to post: [what needs to happen first]

**Pre-flight checklist:**
- [ ] The post links to a real artefact or result
- [ ] The hook survives a two-second skim
- [ ] The CTA is explicit
- [ ] The claims match the evidence
- [ ] The platform and timing are deliberate

## Evidence
Source: the rubric synthesizes standard pre-publish review criteria — strategic alignment, proof of work, content quality, timing, and risk — normalized to 100 points with a 70-point POST threshold. The weights are a starting point; calibrate them against your own post history.

## Examples
**Draft:** "I'm building an open-source content engine. Coming soon."
**Verdict:** DON'T POST (48/100). No artefact, no number, no CTA. Announces intent instead of demonstrating work.
**Fix:** Publish a short case study of something already shipped, with a link and one measurable result, then post that.

## Metrics
- Verdict calibration: fraction of POST recommendations that hold up after publishing. Target: 70%+.
- Score distribution across evaluated drafts. Target: fewer than 20% of drafts below 50/100 after revision.
- Time from draft to decision. Target: under 10 minutes.

## Anti-Patterns
- Posting "I'm building" without showing what was built
- Posting the same content on X and LinkedIn without adaptation
- Posting without a CTA
- Posting when there is nothing to link to
- Posting too frequently without substance
- Posting about AI without demonstrating AI expertise
- Treating the score as a substitute for audience judgment
