agentleFS
stepaniniwisstepaniniwis/claude-skills/skills/audit/saturation-auditor/SKILL.md

saturation-auditor

Skill0 starsChanged 7 months agoMITView on GitHub

Expert skill for defining, assessing, and reporting saturation in qualitative research — grounded in Moore, Aguinis & Darden (2026) but applied pragmatically. Two modes: AUDIT (diagnoses weaknesses in existing saturation sections) and DRAFT (writes saturation content from scratch for methods or results sections). Covers all qualitative designs: Gioia, interview-based, grounded theory, mixed methods, ethnography, case studies. Outputs APA 7th prose aligned with A-journal standards. Trigger on: "saturation section", "how many interviews", "did I reach saturation", "write my saturation", "is my saturation defensible", "check my methods section for saturation", "Sättigung", "Theoretische Sättigung", "sample size justification qualitative", "audit my saturation", or whenever a qualitative paper draft is presented and saturation is at stake.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    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:

    Check the text against the 8 Decision Points from Moore et al.:

    line 74
  2. Distinctive:

    If user wants it, produce revised prose for the weakest sections only.

    line 104
  3. Distinctive:

    Do not rewrite the entire methods section unless explicitly asked.

    line 105
  4. Distinctive:

    Write the saturation content as publication-ready APA 7th prose.

    line 140
  5. Distinctive:

    When and why saturation was declared (or: why the final sample is sufficient given constraints, if full saturation wasn't achieved)

    line 147
  6. Distinctive:

    Use in-text as: (Moore et al., 2026)

    line 184
  7. Distinctive:

    Always apply anti-ai-writing principles: no hedging boilerplate, no AI tells.

    line 195

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “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

Saturation Auditor

Core Reference

Moore, O. A., Aguinis, H., & Darden, T. R. (2026). Defining, assessing, and reporting saturation in qualitative research: Review and recommendations. The Leadership Quarterly. https://doi.org/10.1016/j.leaqua.2026.101950

The Moore et al. framework is the backbone — but always apply with judgment. A-journal reviewers expect fluency in this literature. Use it to show rigor, not to produce bureaucratic checklists.


Mode Detection

Before doing anything, determine which mode applies:

InputMode
User pastes existing methods/results text→ AUDIT
User describes their study and asks for help writing saturation→ DRAFT
User asks "is this enough?" with sample info→ AUDIT
User asks "how do I justify my sample size?"→ DRAFT
Both text AND request to improve/write→ AUDIT first, then DRAFT the improved version

Conceptual Foundation (internalize, don't output verbatim)

Moore et al. distinguish four saturation types. Map the study to one before proceeding:

TypeUse when...Common in...
Theoretical saturationBuilding theory inductively from dataGrounded theory, Gioia methodology
Inductive thematic saturationExploring patterns without formal theorizingThematic analysis, exploratory interview studies
A priori thematic saturationTesting/verifying predefined codesPost-positivist, deductive qualitative, mixed methods
Data saturationPragmatic endpoint — no new infoLarge studies, resource-constrained, mixed methods qual phase

The three failure modes (from their literature review of 429 sources):

  • 42% used saturation without defining it
  • 59.5% gave no assessment methodology
  • 76.3% claimed saturation with zero supporting evidence

Your job is to make sure neither of these is true of the text you audit or write.


MODE 1: AUDIT

When to use

User has an existing methods section, dissertation chapter, or draft paper.

Procedure

Step 1 — Quick diagnostic (output this as a table)

Check the text against the 8 Decision Points from Moore et al.:

Decision PointPresent?Quality
DP1: Philosophical stance statedYes/No/PartialNotes
DP2: Saturation type named + justifiedYes/No/PartialNotes
DP3: Initial sample size with flexibility notedYes/No/PartialNotes
DP4: Coding approach describedYes/No/PartialNotes
DP5: Preliminary saturation assessment shownYes/No/PartialNotes
DP6: Credibility checks documentedYes/No/PartialNotes
DP7: Final saturation determination statedYes/No/PartialNotes
DP8: Evidence reported (grid, quotes, table)Yes/No/PartialNotes

Step 2 — Severity assessment

For each gap, classify:

  • 🔴 Desk-reject risk: Missing entirely, would trigger reviewer concern
  • 🟡 R&R territory: Mentioned but too thin to be defensible
  • 🟢 Solid: Adequate for A-journal submission

Step 3 — Prioritized repair list

List max 5 issues in priority order. For each:

  • What's missing/weak
  • Why it matters (link to trustworthiness dimension: transparency, credibility, confirmability, plausibility, or transferability)
  • Exact fix — including suggested language

Step 4 — Optional: Draft repaired text

If user wants it, produce revised prose for the weakest sections only. Do not rewrite the entire methods section unless explicitly asked.


MODE 2: DRAFT

When to use

User needs saturation content written from scratch, or wants a full rewrite.

Procedure

Step 1 — Intake questions (ask only what you don't already know)

Collect:

  1. Research question / phenomenon under study
  2. Methodology (Gioia? Thematic analysis? Grounded theory? Mixed methods?)
  3. Philosophical stance (interpretivist / post-positivist / pragmatic — if known)
  4. Sample: n participants, how recruited, how many interviews/focus groups/cases
  5. How coding was done (who coded, any team-based, what tools)
  6. What happened iteratively: did you add interviews? When did you stop?
  7. Any credibility checks done (member checking, triangulation, negative cases)?
  8. Target section: Methods only? Results? Both? Word budget?

Step 2 — Map to saturation type

Based on intake, determine the correct saturation conceptualization. State your reasoning in one sentence before drafting (helps the user catch mismatches early).

Example: "Your Gioia design with inductive coding and theoretical categories maps to theoretical saturation — the stopping criterion is theoretical, not just informational redundancy."

Step 3 — Draft

Write the saturation content as publication-ready APA 7th prose. Structure the draft to cover — in natural paragraph flow, not as headers:

  1. The saturation type and its philosophical alignment with the study's paradigm
  2. Initial sample size rationale + iterative expansion logic
  3. How coding tracked emergent themes (the assessment mechanism)
  4. Credibility checks employed
  5. When and why saturation was declared (or: why the final sample is sufficient given constraints, if full saturation wasn't achieved)

Tone and length norms by target:

  • PhD chapter: ~300–600 words, full elaboration, can cite Moore et al. directly
  • Journal article methods section: ~150–300 words, embedded, no methodological footnotes unless the journal is a methods special issue
  • Response to reviewer: Match the reviewer's specific concern, quote their language back, ~100–200 words

Anti-patterns to avoid in both modes

Never write:

  • "Saturation was reached after N interviews" — without saying how you know
  • "Following Glaser and Strauss (1967)" — without specifying which conceptualization
  • "No new themes emerged" — without showing the coding/assessment mechanism
  • Vague hedges like "sufficient data were collected" — always specify the criterion

Watch for the False Positive trap (Moore et al., Fig. 1, Cell 3): Premature saturation claims are the most common error. If the study has fewer than ~12 interviews with no iterative assessment, flag this explicitly.

Code saturation ≠ meaning saturation (Hennink et al., 2017): New codes stopping is not the same as understanding the phenomenon fully. Especially in Gioia-style work, meaning saturation is what matters.


APA 7th Citation for Moore et al.

Moore, O. A., Aguinis, H., & Darden, T. R. (2026). Defining, assessing, and reporting saturation in qualitative research: Review and recommendations. The Leadership Quarterly. Advance online publication. https://doi.org/10.1016/j.leaqua.2026.101950

Use in-text as: (Moore et al., 2026)


Output principles

  • Prose only. No bullet-point saturation sections in drafted text.
  • APA 7th in-text citations where relevant.
  • Match the voice of the surrounding text if user provides context.
  • Flag philosophical inconsistencies (e.g., claiming inductive discovery while using fully predetermined codes) — these are reviewer magnets.
  • Always apply anti-ai-writing principles: no hedging boilerplate, no AI tells.

Download now (ZIP)

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.