Content Quality Check
The QA gate a draft passes before it publishes. It exists because of one fact:
readers now pattern-match AI tells in seconds, and every tell that ships spends the brand's
credibility. The check exists to catch what the writer can no longer see in their own draft.
It reports before it rewrites. The deliverable is a severity-tagged review with a fix per flag;
wholesale rewriting happens only when asked, because the writer owns the draft.
What this consumes
| Input |
Why |
If missing |
| The draft |
The thing under review |
Required |
| Channel and audience |
Calibrates length, register, and which rules bind hardest |
Ask in one line |
| Voice directory (optional) |
brand-voice-guide output: the pillars to check conformance against and the Avoid list the linter enforces |
Run with the universal checklist only, and say so |
The division of labor across the repo holds: brand-voice-guide builds the standard, this skill
enforces it on drafts, and outbound-engine carries its own sequence-specific gate (this skill
can still review sequence copy as prose).
The workflow
Four phases. Mechanical first, judgment second, so the model's attention goes where the machine
cannot.
Phase 1: Intake and calibration
Read the draft, note channel and audience, and locate the voice directory if one exists (pillars
from voice-core.md, banned words from the Avoid section of writing-standard.md). State the
bar being applied: universal checklist only, or universal plus brand standard.
Phase 2: Mechanical sweep
Run scripts/check_content.py on the draft, with --avoid pointing at the brand's Avoid words
when available. The linter catches everything a machine can catch deterministically: slop phrases,
LLM-overrepresented words, "not just X, it's Y" reframes, "let's dive in" openers, announcement
cliches, em-dash chaining, rhetorical-question openers, empty-intensifier density, weak-closer
formulas, and the brand's own banned words. Same draft, same flags, every run.
Phase 3: Editorial judgment pass
Read the draft as an editor, against references/detection-checklist.md section C, for what no
regex can see:
- The opener: does the first line earn attention, or does it throat-clear and stage-set?
- The closer: does it end flat and forward, or restate the opening as a summary?
- Proof: does every claim carry a number, a name, or evidence, or does it float? Flag
unproven claims as "needs proof"; verifying truth is the writer's job, naming the gap is ours.
- Specificity: concrete nouns and real examples over abstractions ("the 300-row
questionnaire" beats "operational inefficiencies").
- Template shape: does the structure read as a filled-in listicle or a piece with an argument?
- Voice conformance (when a directory exists): would this pass the brand's distinctiveness
test, and does each section sound like the pillars or merely avoid the banned words?
Phase 4: Report, then revise on request
Deliver the review: verdict (publish / revise / rework), the linter output verbatim, judgment
findings with line references, and a concrete fix per flag. If the user asks for the revision,
apply the fixes, re-run the linter until clean, and hand back a before/after so the changes are
inspectable. The final gate is mechanical PASS plus an editor's clear.
Scripts
| Script |
What it does |
Network |
Keys |
scripts/check_content.py |
Deterministic content linter: universal tells plus the brand's Avoid list, severity-tagged with line numbers |
None |
None |
Failure modes to avoid (the quality bar)
- Silent rewriting. The review comes first; the writer decides what to accept.
- Flag-counting without fixes. Every flag ships with its fix, or it is nagging, not QA.
- Enforcing taste as rule. The checklist distinguishes severities; a LOW is advisory and says
so. Real lists of three real things are not tricolon abuse.
- Universal-only when a standard exists. If the brand has a voice directory, checking against
the generic list alone is a half-done job.
- Missing the structural tells. A draft can pass every word-level check and still read as a
template; the judgment pass exists for exactly that.
- Grading truth. "Needs proof" is a finding; fact-checking the proof is out of scope and
said so.
Bundled references
references/detection-checklist.md: the full catalog, severity-tagged: A (mechanical, the
linter's spec), B (word and phrase lists), C (judgment checks). Read at Phases 2-3.examples/cohere/: a deliberately slop-ridden Cohere announcement draft, the review that
catches 20+ flags across both passes, and the revised draft in Cohere's actual voice that the
linter passes clean.
1---2name: content-quality-check3description: Editorial QA pass that catches AI tells, template language, hollow hedges, weak closers, and brand-voice violations in any draft before it publishes. Runs a deterministic linter (slop phrases, LLM-overrepresented words, "not just X, it's Y" reframes, em-dash chaining, rhetorical-question openers, announcement cliches, the brand's own Avoid list) and then an editorial judgment pass (throat-clearing openers, closers that restate the open, claims without proof, specificity, voice-pillar conformance). Reports findings by severity with line references and a fix per flag; rewrites only on request. Use whenever the user says "review this draft", "QA this post", "check for AI tells", "does this sound AI-written", "editorial review", "content quality check", "de-slop this", or is about to publish copy. Enforces the standard that brand-voice-guide builds; works standalone with the universal checklist when no voice directory exists.4---56# Content Quality Check78The QA gate a draft passes before it publishes. It exists because of one fact:9**readers now pattern-match AI tells in seconds, and every tell that ships spends the brand's10credibility.** The check exists to catch what the writer can no longer see in their own draft.1112It reports before it rewrites. The deliverable is a severity-tagged review with a fix per flag;13wholesale rewriting happens only when asked, because the writer owns the draft.1415## What this consumes1617| Input | Why | If missing |18|---|---|---|19| The draft | The thing under review | Required |20| Channel and audience | Calibrates length, register, and which rules bind hardest | Ask in one line |21| Voice directory (optional) | `brand-voice-guide` output: the pillars to check conformance against and the Avoid list the linter enforces | Run with the universal checklist only, and say so |2223The division of labor across the repo holds: `brand-voice-guide` builds the standard, this skill24enforces it on drafts, and `outbound-engine` carries its own sequence-specific gate (this skill25can still review sequence copy as prose).2627## The workflow2829Four phases. Mechanical first, judgment second, so the model's attention goes where the machine30cannot.3132### Phase 1: Intake and calibration3334Read the draft, note channel and audience, and locate the voice directory if one exists (pillars35from `voice-core.md`, banned words from the Avoid section of `writing-standard.md`). State the36bar being applied: universal checklist only, or universal plus brand standard.3738### Phase 2: Mechanical sweep3940Run `scripts/check_content.py` on the draft, with `--avoid` pointing at the brand's Avoid words41when available. The linter catches everything a machine can catch deterministically: slop phrases,42LLM-overrepresented words, "not just X, it's Y" reframes, "let's dive in" openers, announcement43cliches, em-dash chaining, rhetorical-question openers, empty-intensifier density, weak-closer44formulas, and the brand's own banned words. Same draft, same flags, every run.4546### Phase 3: Editorial judgment pass4748Read the draft as an editor, against `references/detection-checklist.md` section C, for what no49regex can see:5051- **The opener:** does the first line earn attention, or does it throat-clear and stage-set?52- **The closer:** does it end flat and forward, or restate the opening as a summary?53- **Proof:** does every claim carry a number, a name, or evidence, or does it float? Flag54 unproven claims as "needs proof"; verifying truth is the writer's job, naming the gap is ours.55- **Specificity:** concrete nouns and real examples over abstractions ("the 300-row56 questionnaire" beats "operational inefficiencies").57- **Template shape:** does the structure read as a filled-in listicle or a piece with an argument?58- **Voice conformance** (when a directory exists): would this pass the brand's distinctiveness59 test, and does each section sound like the pillars or merely avoid the banned words?6061### Phase 4: Report, then revise on request6263Deliver the review: verdict (**publish / revise / rework**), the linter output verbatim, judgment64findings with line references, and a concrete fix per flag. If the user asks for the revision,65apply the fixes, re-run the linter until clean, and hand back a before/after so the changes are66inspectable. The final gate is mechanical PASS plus an editor's clear.6768## Scripts6970| Script | What it does | Network | Keys |71|---|---|---|---|72| `scripts/check_content.py` | Deterministic content linter: universal tells plus the brand's Avoid list, severity-tagged with line numbers | None | None |7374## Failure modes to avoid (the quality bar)7576- **Silent rewriting.** The review comes first; the writer decides what to accept.77- **Flag-counting without fixes.** Every flag ships with its fix, or it is nagging, not QA.78- **Enforcing taste as rule.** The checklist distinguishes severities; a LOW is advisory and says79 so. Real lists of three real things are not tricolon abuse.80- **Universal-only when a standard exists.** If the brand has a voice directory, checking against81 the generic list alone is a half-done job.82- **Missing the structural tells.** A draft can pass every word-level check and still read as a83 template; the judgment pass exists for exactly that.84- **Grading truth.** "Needs proof" is a finding; fact-checking the proof is out of scope and85 said so.8687## Bundled references8889- `references/detection-checklist.md`: the full catalog, severity-tagged: A (mechanical, the90 linter's spec), B (word and phrase lists), C (judgment checks). Read at Phases 2-3.91- `examples/cohere/`: a deliberately slop-ridden Cohere announcement draft, the review that92 catches 20+ flags across both passes, and the revised draft in Cohere's actual voice that the93 linter passes clean.
esmefong