Job Application Board — 3-Lens Resume Review

Review a resume and cover letter against a job posting for ATS keyword misses, generic recruiter appeal and weak interview hooks before you submit.

How it works

Runs a three-phase board review. Phase 1: three reviewer lenses (ATS Parser Realist, Recruiter Skeptic, Interview Predictor) each flag issues one at a time, quoting the actual resume or cover letter text and stating the concrete consequence. Phase 2: each lens challenges and corroborates the other two lenses' findings. Phase 3: one verdict tags every finding CONFIRMED, CORROBORATED or DISPUTED (challenged findings stay listed as DISPUTED) and ends with a single Submit recommendation. It is deliberately not encouraging. Findings are grounded in the text you paste, and missing requirements are flagged instead of guessed. If no job posting is given, ATS findings are best-effort against a generic posting and it asks for the real one. It treats contact details as context and does not echo them; the pasted text is sent to the model as-is, so redact what you do not want shared.

Sample output

Sample input (an invented example):

Use the Job Application Board on this (a sample I wrote). Please review my resume and cover letter against the posting before I submit.

POSTING (Data Analyst, Example Retail Co.): Requires SQL, Tableau, A/B testing, stakeholder reporting, 2+ years experience. Nice to have: Python, dbt.

RESUME (Sam Rivera, [email protected]):
Summary: Results-driven professional with a passion for data.
Experience:
- Business Associate, Sample Goods Ltd, 2022-2024: Responsible for weekly reports. Worked with the sales team on dashboards in Power BI. Helped improve the checkout process.
- Intern, Demo Logistics, 2021: Cleaned spreadsheets and queried a database with SELECT statements.
Skills: Excel, Power BI, communication, teamwork, data analysis.
Gap: 2024 to present (no explanation).

COVER LETTER:
"Dear Hiring Manager, I am writing to apply for the position advertised. I am a hard worker and fast learner who would be a great fit for your company. I look forward to hearing from you."

Real, unedited output:

## Phase 1 — Lens findings
Mode: one lens at a time

### 🤖 ATS Parser Realist
- **A1. "SQL" never appears.** The only evidence is "queried a database with SELECT statements." A search for "SQL" won't surface this resume. *Fix:* add "SQL" to Skills and the intern bullet.
- **A2. "Tableau" never appears.** The only BI tool listed is Power BI ("dashboards in Power BI"). *Fix:* add Tableau only if you've actually used it. Otherwise, list Power BI as your BI tool and accept the keyword miss.
- **A3. "A/B testing" never appears.** "Helped improve the checkout process" is the only place it could fit. *Fix:* if that work involved a test, say so with the term "A/B test."
- **A4. "Stakeholder reporting" never appears.** The closest lines are "Responsible for weekly reports" and "Worked with the sales team." *Fix:* reword to "weekly stakeholder reporting for the sales team."
- **A5. Nice-to-haves "Python" and "dbt" are absent.**
- **A6. No title matches "Data Analyst."** The titles are "Business Associate" and "Intern," and the Summary doesn't name the target role either. *Fix:* put "Data Analyst" in the Summary headline. Don't change the real job title.
- **Top-keyword check:** 0 of the 4 required skill terms (SQL, Tableau, A/B testing, stakeholder reporting) appear verbatim.

### 🧑‍💼 Recruiter Skeptic
- **R1. The Summary is pure template:** "Results-driven professional with a passion for data." It says nothing in the first 10 seconds.
- **R2. The bullets describe duties, not outcomes:** "Responsible for weekly reports," "Helped improve the checkout process." There's nothing to remember.
- **R3. "2024 to present" is unexplained.** An unexplained gap gets read as a red flag by default.
- **R4. The cover letter could go to any employer:** "the position advertised," "your company." It doesn't name Example Retail Co., the Data Analyst role, or any requirement from the posting.
- **R5. The experience threshold is ambiguous.** Dates are year-only. 2022–2024 is 2 years by subtraction (2024 − 2022 = 2), but it could be anywhere from 1 to 3 years depending on the months.

### 🎯 Interview Predictor
- **I1. Claims with no evidence:** "Results-driven," "passion for data," "hard worker and fast learner." None of them gives an interviewer something to ask about.
- **I2. "Helped improve the checkout process" is the best hook, but it has no metric.** This work clearly had a measurable outcome (conversion, drop-off, time), and none is given. It's also unclear what "helped" means you personally did.
- **I3. No answer to "why this role, why this company"** anywhere in the materials.
- **I4. The Power BI dashboards have no story.** We don't know who used them, what decision they changed, or how often they were used. That is exactly the "stakeholder reporting" question an interviewer would ask.
- **I5. "Communication, teamwork" are listed as skills with no supporting bullet.**

## Phase 2 — Cross-examination
- **[A1 SQL]** Recruiter: not a false positive; SQL is the first requirement listed. Interview **corroborates**: "SELECT statements" invites a "how deep is your SQL?" question, and nothing here answers it.
- **[A2 Tableau]** Recruiter **corroborates**: a skimmer checking requirements sees Power BI, not Tableau. Interview: Power BI is transferable, but that softens the issue rather than removing it, so this is not a substantiated challenge.
- **[A3 A/B testing]** Recruiter: no challenge. Interview **corroborates**: the checkout work is the natural place for a test story, if one exists.
- **[A4 Stakeholder reporting]** Recruiter: no challenge. Interview **corroborates**: this is the same gap as I4.
- **[A5 Python/dbt]** Recruiter **challenges**: "These are nice-to-haves, so they won't filter out a candidate who meets the requirements." Interview: no corroboration.
- **[A6 Title]** Recruiter: minor, but not a false positive. Interview: no view. No corroboration.
- **[R1 Generic Summary]** ATS **corroborates**: the Summary contains zero posting keywords. Interview **corroborates**: this is the same issue as I1/I5.
- **[R2 Duty bullets]** ATS: no challenge. Interview **corroborates**: this is the same issue as I2.
- **[R3 Gap]** ATS: parsers don't penalize gaps, but that doesn't make the human risk false. Interview **corroborates**: it will be asked, and there's no prepared framing.
- **[R4 Generic cover letter]** ATS **corroborates**: the cover letter contains zero posting keywords. Interview **corroborates**: this is the same issue as I3.
- **[R5 Experience ambiguity]** ATS **challenges**: "2024 − 2022 = 2 meets '2+ years' on its face, and the 2021 internship adds more." Interview: no corroboration.
- **[I1/I5 Unsupported claims]** ATS: soft-skill words aren't posting keywords, so they add nothing. Recruiter **corroborates** (R1).
- **[I2 No metric]** Recruiter **corroborates** (R2).
- **[I3 No "why"]** Recruiter **corroborates** (R4).
- **[I4 Dashboard story]** ATS **corroborates** (A4).

## Phase 3 — Board verdict
1. **[CORROBORATED]** "SQL" is missing as a keyword. The only evidence is "SELECT statements." — *fix:* add "SQL" to Skills and the intern bullet, naming what you actually did (joins, aggregations, etc.).
2. **[CORROBORATED]** Tableau, a hard requirement, is absent. Only Power BI is listed. — *fix:* add it only if true. Otherwise, frame Power BI as your BI tool and expect this to cost you some searches.
3. **[CORROBORATED]** "A/B testing" is absent. — *fix:* if the checkout work involved a test, rewrite that bullet around it. Don't add the term if it didn't.
4. **[CORROBORATED]** There's no "stakeholder reporting" keyword and no dashboard story (A4 + I4). — *fix:* e.g. "Built weekly Power BI stakeholder reports for the sales team, used for [decision]." Use only real details.
5. **[CORROBORATED]** The Summary and Skills are generic, unsupported claims (R1 + I1 + I5). — *fix:* replace them with "Data Analyst with [N] years of SQL/Power BI reporting…" and cut "communication, teamwork" unless a bullet backs them up.
6. **[CORROBORATED]** The bullets list duties, and the checkout win has no metric (R2 + I2). — *fix:* "Analyzed checkout data → [change] → [X% result]," using your real numbers.
7. **[CORROBORATED]** The "2024 to present" gap is unexplained. How long it is depends on when "present" is. *Assumption:* if you submit in late 2026, it's roughly 2 years (2026 − 2024). — *fix:* add a one-line reason on the resume or in the cover letter, plus any upskilling you did.
8. **[CORROBORATED]** The cover letter is generic, has no keywords, and gives no "why Example Retail Co." (R4 + I3). — *fix:* name the role and company, match 2–3 posting requirements to specific evidence, and give one real reason for this company.
9. **[CONFIRMED]** No "Data Analyst" title appears anywhere. — *fix:* use it as the Summary headline.
10. **[DISPUTED]** The 2+ years requirement is ambiguous because dates are year-only. — *challenge:* "2024 − 2022 = 2 already meets the threshold, plus the 2021 internship." *(Your call: adding months removes the doubt.)*
11. **[DISPUTED]** Python and dbt are absent. — *challenge:* "They're nice-to-haves, not filters." *(Your call: add them only if you have real experience.)*

**Submit recommendation:** Fix first. There are 8 CORROBORATED findings (3 of the 4 required skill keywords, the stakeholder-reporting gap, the generic summary, unquantified bullets, the unexplained gap, and a generic cover letter) and 1 CONFIRMED title gap. Rewrite the resume's Summary and Experience sections and the full cover letter before applying. The 2 DISPUTED items are your call.

Report consistency rules

• The first line under the Phase 1 heading must be Mode: one lens at a time (or Mode: sub-agents if sub-agents were actually used). Never omit it.
• Tags follow Phase 2 exactly: a finding another lens corroborated is CORROBORATED (never CONFIRMED); a finding with a concrete challenge is DISPUTED; CONFIRMED means no lens challenged or corroborated it. One issue gets one tag: merge duplicates.
• Every Phase 1 finding must either appear in Phase 3 or be merged into a listed item; list tagged items in the order CORROBORATED, CONFIRMED, DISPUTED.
• Every count in the closing line must equal the number of tagged items above it; recount before writing it.
• Use only facts, figures and dates that are in the pasted input. Anything else (market rates, outside examples, assumptions about the sender, reader or timing) must be labelled "assumption" or "unverified", never stated as fact. When comparing dates or durations, state the reference date used and write out the arithmetic.
• Cite clause, line or item numbers only from the input text, and check the number against the text before citing it.
• Do not use today's date or assume any date has passed unless the input states a current date; if a point depends on timing, write it as "if X, then ..." and label it an assumption.
• Count requirements, keywords or items only as they are listed in the input (recount the list before writing a number such as "0 of N").

Handling sensitive info in the application

If the resume or cover letter contains contact details, references, or other personal information,
treat that as context only — do not echo it back beyond what's needed to explain a finding. This
Skill sends the text you paste as-is to the model; redact anything you'd rather not share before
pasting.

Works with

Built for Claude Code; other agents that read SKILL.md files (such as Codex CLI or OpenClaw) should also work. No network access or extra tools needed.

Product and company names mentioned here, including Anthropic, OpenAI and OpenClaw, are trademarks of their respective owners; this Skill is an independent product and is not affiliated with, endorsed by or sponsored by any of them.