Job Post Desk: agent fixes for your job ads and applicant replies, pay range shown, fair wording, owner approved
by PaperChaseWebb
Let an agent tidy your job ads, check where each ad is still up and review draft replies to applicants, without dropping the pay range, inventing a bonus, asking for pay history or telling anyone they got the job. It writes fixed copies and never posts, takes down, rejects or replies. A fix check refuses a removed range, a deleted line, a new perk or a new email. You approve at your terminal and post it yourself.
- Price
- $29
- Difficulty
- Intermediate
- Setup
- 45 min
- Format
- .md file
job-post-desk-pay-range-shown-fair-wording-owner-approved.md · 207 KB · instant download after checkout
Want this built for you instead? We can build it for you.
What you end up with
Each job ad shows the range you set, describes benefits, gives a close date and one way to apply, and drops pay history questions and age, gender, health and origin wording; you know which board ads are stale or show the wrong pay, one approved change at a time, with nothing deleted and no claim that anything is compliant.
What you get
- 21 sections and an appendix in one .md file, with a one-line unpack for all 73 files
- jobpost_check.py posts: 21 checks on each job ad against your own settings and a dated state table, from no range or "up to $45 DOE" and no benefits or close date to pay history questions, "digital natives", "He must", "able-bodied", an unpaid trial day and no software screening notice
- jobpost_check.py postings: 9 checks on where each ad is up, so you see ads still up after the role was filled or the close date, boards with no pay or the wrong range and duplicates
- jobpost_check.py replies: 13 checks on draft replies to applicants, such as pay history, age or family questions, a fee, an ID number before any offer, a job offer the owner did not make and "our system rejected you"
- jobpost_check.py fix: 12 checks that compare the agent's copy with the original, so a deleted line, a removed range or apply link, a new perk, email or figure fails
- A claims check that stops reports saying an ad is compliant or that the agent posted, hired, rejected or replied
- The owner step: approve.py needs a real terminal and part of the file hash, refuses agent sessions and piped input, and checks the original has not changed before you mark it posted
- An eval set for a made-up dental group with 3 ads, a postings list and 12 reply threads, rules and two skills for Grok Bot, Claude Code, Codex and Hermes, six prompts and templates
- The test harness: 62 unit tests and 779 planted faults you can repeat
Who it is for
Small businesses and agencies who want an agent to tidy job ads and draft applicant replies, with a person approving every change and making every hiring decision.
Works with
Solid chips: the agent is named in this listing's own tools list. Dashed chip: Plain Markdown, so it also pastes into other agents such as Grok Bot, Hermes, OpenClaw or Muse. We list those as compatible by format only, not as tested.
Tools needed
Cost to run: The kit is the only purchase. Everything runs locally with Python; the checker makes no network calls, logs in to nothing and changes nothing on any job board.
Time to result: The worked example in 5 minutes; your own ads and boards checked in about an hour
How it works
Unpack and test
Save the kit as pack.md in an empty folder, run the unpack line from Appendix A, then sh test_all.sh.
Expected: 73 files written; 62 tests OK; 779 of 779 planted faults caught.
Watch the flow
Run sh example/run_example.sh and read it next to section 17.
Expected: A plumbing company's three ads with 23 findings, a postings list with 7, a bad reply with 8, the agent's first-try fixes caught with 11, the second tries clean, a first-try report with 5 findings and a piped approval refused.
Put in your roles
Fill in site/settings.json and site/facts.md from the templates, paste your ads into site/jobs/, list your boards in site/postings.csv, then run python3 desk/jobpost_check.py site and python3 desk/jobpost_check.py posts.
Expected: A short list of settings and ad lines to fix first.
Fix and sweep the boards
Have your agent follow skills/job-post-fix and skills/applicant-reply-review, then run jobpost_check.py fix queue/fixes and approve each file at your terminal.
Expected: Fixed ads in queue/fixes with 0 high findings, and a list of board ads to take down or correct yourself.
Sample prompts, free to use
Follow skills/job-post-fix for every post in site/jobs. Do not post anything. List the codes fixed and the codes left for me, with the reason for each.
Prompt 1 of 6.
Follow skills/applicant-reply-review for site/postings.csv. Take nothing down. List each board, the posting id and what I should do there: take it down, fix the pay or add benefits.
Prompt 2 of 6.
Run jobpost_check.py posts and replies. For each J10, J19 and C06, tell me the benefit, perk or number so I can add it to facts.md or take it out.
Prompt 4 of 6.
Free preview of the file
# Job Post Desk: agent fixes for your job ads and applicant replies, pay range shown, fair wording, owner approved Version 1.0, 2026-11-01. By PaperChaseWebb. For small businesses that hire (trades, clinics, restaurants, shops, offices, home services) and the agencies that run their hiring pages, who want an AI agent to tidy job ads, keep track of where each ad is up and draft replies to applicants, without inventing pay, a bonus or a perk, asking what nobody should ask, or telling an applicant they got the job. Works with Grok Bot, Claude Code, Codex and Hermes. Python 3.8 or newer, standard library only, no API keys, no logins. The kit and the agent never post or take down an ad, never decide on an applicant and never send anything. Tested with 62 unit tests and 779 planted faults on Linux with Python 3.13, 3.12, 3.11, 3.9 and 3.8. All business, contact and applicant data in this kit is made up. Not legal advice. Job ads now carry rules. In the table this kit reads, 14 states and DC ask covered employers to put a pay range in job posts, and Delaware joins in September 2027. Colorado, Connecticut, Illinois, Maryland, Minnesota, New Jersey and Washington also ask for a general description of benefits, and Colorado asks for the date applications close. Many states and cities bar asking for pay history. New York City and Illinois ask employers to tell applicants when software is used to screen them. Federal rules limit ads that show a preference by age, sex, religion, nationa
Test record
- Date
- Nov 1, 2026
- What was run
- tests/test_lib.py (8), test_site.py (3), test_posts.py (12), test_postings.py (7), test_replies.py (6), test_fix.py (8), test_claims.py (2), test_modes.py (6), test_mini.py (2) and test_approve.py (8, on a real pseudo terminal); tests/mutate.py with 713 planted faults across posts, postings lists, replies, bad fixes, reports, agent rules and settings, plus 66 against jobpost_mini.py; jobpost_check.py set on the eval set; example/run_example.sh end to end; and the unpack line with test_all.sh. Linux, Python 3.13, 3.12, 3.11, 3.9 and 3.8.
- Result
- 62 passed, 0 failed on all five versions; 779 of 779 planted faults caught and all clean baselines clean; the eval set matched its 35 expected codes and every fixed ad had 0 high findings; the example caught the first-try fixes with 11 findings and the first-try report with 5, and refused piped approval; the unpack line wrote every file and test_all.sh passed there.
- Not run
- A live Grok Bot, Claude Code, Codex or Hermes session. Job boards and applicant tracking systems. City rules other than the NYC software notice, and laws outside the US. Languages other than English. macOS or Windows. Legal review.
Copied from this item's dated test notes. What was not run stays in.
Worth knowing
- The agent rules were checked by the checker but not loaded inside a live Grok Bot, Claude Code, Codex or Hermes session for this release. The eval fixes were written by hand to the rules. Give your agent the entrance exam in section 15 before you rely on it.
- Not legal advice. Nothing in the kit says an ad, a reply or a business meets any law. The state table is dated 2026-11-01; check it against the sources in section 20. City rules other than the NYC software notice and laws outside the US were not checked, and every state a remote role hires from is treated as asking.
- The checker reads the ads, postings list and replies you save. It does not read job boards or applicant tracking systems and cannot post or take anything down, so open each live ad on a phone after each change. English only.
- The owner step assumes the agent cannot run approve.py at a terminal. Run it as a separate user; agent rules alone are not a hard boundary.
- All businesses, people, roles and numbers in the kit are made up. Tested on Linux only; macOS and Windows were not tested.
- On 2026-11-01 on Linux: 62 unit tests and 779 of 779 planted faults passed on Python 3.13, 3.12, 3.11, 3.9 and 3.8; the eval set matched its 35 expected codes; the unpack line wrote every file and test_all.sh passed in that folder.
Questions
Which agents does it work with?
Any agent that can read files and run a script. The kit has rules for AGENTS.md and CLAUDE.md, Claude Code and Codex notes, Hermes notes and a Grok Bot standing order. Those were checked by the checker, not loaded in each agent for this release.
Will it post ads, reject applicants or reply for me?
No. The agent writes fixed copies into a queue folder and a list of board ads to correct. You approve each file at a terminal, post it yourself, and make and send every hiring decision.
Which job boards does it work with?
Any board or careers page you can copy an ad from. You list where each ad is up in a CSV; the kit never logs in anywhere.
Can I keep a record check or a citizen requirement?
Yes, if the role needs it and your settings say so. A citizen line passes when the role is marked as needing it; a blanket record screen is a medium finding left for you to decide.
Can you set it up for us?
Yes. Our Install sprint maps your roles, ranges and boards, writes your facts file and builds an eval set from your own ads, or Managed Ops runs the checks, fixes and board sweeps for you each week. Book a call or write to [email protected].
Details
Version 1.0 · Updated Nov 1, 2026 · Markdown version of this page
License: Use inside your own business or for your clients. Do not resell or republish the kit itself. Download terms
Related
Job Post Desk: fair job ad rules and a ten-check job ad scan (free sample)
From Job Post Desk: the ways small job ads slip, ten rules to paste into your agent, and a read-only script that runs ten checks on one job ad, from no pay range or "up to $45 DOE" and no benefits or way to apply to pay history questions, age-coded or gendered words and a compliance claim.
Refund and Returns Desk: agent fixes for your return policy, delivery page and delay notice, policy matched, refund amounts checked, owner approved
Let an agent tidy your return policy, delivery page, product page and delay notice, check refunds and late orders against your order list, and review replies to refund requests, without inventing a window, a fee or free returns, or telling a customer a refund was made when it was not. It writes fixed copies and never touches the live site, refunds anything or sends anything. A fix check refuses a removed return window, a deleted sentence, a new promise or a new email. You approve at your terminal and paste it in yourself.
Review Reply Desk: AI-drafted review replies, checked for privacy and offers, approved by the owner
Let an agent draft your Google and Yelp review replies without the mistakes that cost money. A checker refuses drafts that use a last name or details from your records, offer a discount for a better review, ask for a review to be changed, argue, invent numbers or promises, or repeat your last reply word for word. Health topics, allergies and legal threats go to the owner, clinics get a stricter privacy mode, and nothing is posted until the owner approves it at a real terminal.
Email Sender Fix Desk: agent fixes for one-click unsubscribe, footers and SPF, DKIM and DMARC, checked against the original, owner approved
Let an agent put the one-click unsubscribe headers, a visible unsubscribe link and your postal address into your newsletters, and propose clean SPF, DKIM and DMARC records, the things Gmail, Yahoo and Outlook ask senders for. It writes fixed copies and never sends or touches DNS. A fix check refuses deleted words, made-up addresses, tracking pixels and hidden links. A records check refuses dropped senders, invented keys and DMARC jumped straight to reject. You approve at your terminal and make the change yourself.
Chat Assistant Disclosure Desk: agent fixes for your website chat, AI said up front, a person on request, owner approved
Let an agent tidy your website chat assistant's greeting, instructions, handoff text and AI use page, and review a week of saved chats, without the assistant passing as a person, keeping visitors from a person, or making up prices and promises. It writes fixed copies and never touches the live chat. A fix check refuses a removed AI line, a deleted emergency line, a new discount or a new phone number. A chat check flags every reply that claimed to be human, ignored a request for a person or quoted a price you never wrote down. You approve at your terminal and paste it in yourself.
Don't Get Sued: a pre-launch legal-risk gate for AI-built apps and agents
A read-only agent gate for the fourteen legal risks that hit small builders most: marketing texts and email, trackers before consent, accessibility, kids' data, biometrics, plain-text personal data, AI disclosure, AI claims you cannot prove, takedowns, policies that do not match the product, renewals and refunds, licenses, and agents that act for users. Evidence rule, gate register, seeded test app, 12 prompts, counsel packet. Not legal advice.
Ask-first rules for AI agents: hard nos, approval cards and a test set
Paste-in rules that make an agent ask before it sends, spends, posts or deletes, define what a real yes is, and stop retries after a no, plus 12 test scenarios to check it.
All-In Price Desk: agent fixes for your menus, price pages and checkout, fees in the first price, owner approved
Let an agent tidy your menus, rate cards, listings and checkout pages, check order records and review draft answers to "what is this fee?", without inventing a price, a discount or a fee, calling your fee a tax or promising a refund. It writes fixed copies and never publishes, refunds, charges or replies. A fix check refuses a removed price line, a deleted fee purpose, a new offer or a new phone number. You approve at your terminal and publish it yourself.