Pursue or Pass bundle: every federal notice gated, scored, decided by you and screened before anyone calls a buyer

What you end up with

Every federal notice you look at in one register, with the gate result, the score, your decision and reason, and proof that the capability statement and any buyer contact were screened, so the hours go to the bids worth making.

What you get

• Four packs in full: Opportunity Triage Kit, Eligibility and Compliance Gate Checklist, Evidence-Only Capability Statement and Government Buyer Outreach Screen ($111 bought separately)
• The Pursuit Desk (only in this bundle): a twenty-column register that carries each notice from found to closed
• The pursuit loop: which pack each stage uses, what it records and what must be true before the next stage
• pursue_check.py: seventeen read-only checks, from a bid after a failed gate to buyer contact without a clean screen
• Owner approval: your agent drafts every field except your name and date on the decision
• Twenty planted-fault tests, a Grok Bot standing order, an AGENTS.md snippet and a skill file
• Eight prompts, a failure table and a fictional worked example that checks clean

Who it is for

Small and veteran-owned firms, and the consultants who help them, who weigh more than a few federal notices a month and want one place that says which ones are worth the hours and why.

Works with

Solid chips: the agent is named in this listing's own tools list. Dashed chip: A Grok Bot standing order, an AGENTS.md snippet for Claude Code and Codex, and a skill file; the checker and tests ran on Linux with Python 3, the agent files were not loaded inside each harness.

What is in the bundle (4 packs)

Tools needed

Cost to run: The bundle is the only purchase. pursue_check.py uses no model tokens; drafting uses whatever your agent normally spends.
Time to result: Your first notice gated, scored and decided in about an hour once the four packs are installed

How it works

1. Run the tests

   Copy the desk files from the bundle and run python3 tests/test_pursue.py.

   Expected: 20 tests pass: the clean example, seventeen planted faults each caught by its own check, a strategic exception and an unreadable file.

2. Install the four packs

   Set up the Triage Kit, Gate Checklist, Capability Statement and Outreach Screen, and run each one's own checks.

   Expected: cap_check.py and outreach_screen.py run from your machine.

3. Start the register

   Copy example/pursuits.csv to govcon/pursuits.csv, keep the header, delete the fictional rows and run pursue_check.py on it.

   Expected: 0 pursuits, 0 findings.

4. Add your agent's rules

   Paste the standing order, AGENTS.md snippet or skill file for your agent from section 6.

   Expected: Your agent drafts rows and leaves decided_by and decided_on empty.

5. Gate, score, decide

   Add your next notice at stage found, have your agent gate and score it, then read the band and type your own decision, name and date.

   Expected: pursue_check.py prints 0 findings, or names the exact gap to close.

Sample prompt from the pack

Walk 01-gate-checklist.md for notice [ID] against our company record. Record PASS or STOP in govcon/pursuits.csv.
On a STOP, name the gate and the rule, set the decision to NO-BID as a draft, and leave decided_by empty for me.

Run pursue_check.py govcon/pursuits.csv --summary.
Tell me which RESHAPE rows come back this month and which no-bids you would take back, with the reason from each row.

Free preview of the file

# Pursue or Pass bundle: every federal notice gated, scored, decided by you and screened before anyone calls a buyer

Version 1.0, 2026-11-03. By PaperChaseWebb. A bundle of four paid GovCon packs (Opportunity Triage Kit, Eligibility and Compliance Gate Checklist, Evidence-Only Capability Statement and Government Buyer Outreach Screen) plus a Pursuit Desk that is only in this bundle. The desk is one register file and a read-only checker with seventeen checks, Python 3 standard library only. Works with Grok Bot, Claude Code, Codex and Hermes through a rules block, a skill file and an AGENTS.md snippet. Tested as described in section 9 on Linux with Python 3. The agent files were not loaded inside each harness for this release, and the desk has not been run on a live solicitation. Not legal advice.

A bid scorecard tells you what a notice is worth. It does not stop you from bidding the one that failed a gate, from overriding a weak score upward because the agency name is exciting, from letting an agent's draft stand in for your decision, or from calling a contracting officer during an open solicitation. Those are the mistakes that cost a small firm a month or its standing with a buyer.

This bundle closes that gap. The four packs each do one stage well. The Pursuit Desk ties them into one register where a notice moves found, gated, scored, decided, outreach, submitted, closed, and a script checks that every row tells a consistent story: the gate result matches the decision, the

Worth knowing

• pursue_check.py checks that each row holds together. It does not read the solicitation and cannot tell you whether you are eligible; the Gate Checklist and the live rules do that. Not legal advice.
• The standing order, AGENTS.md snippet and skill file were not loaded inside Grok Bot, Claude Code, Codex or Hermes for this release, and the desk has not been run on a live solicitation. The worked example is a made-up facilities firm.
• On 2026-11-03, on Linux with Python 3: 20 of 20 desk tests passed, each of 17 planted faults was caught by its own check only, the worked example checked clean, and an unreadable file exited 2. The free sample's 5 checks passed their 6 tests.
• The bundle contains the versions of the four packs listed on this page on the day you buy. Each keeps its own notes, sources and limits.

Details

Version 1.0 · Updated Nov 3, 2026 · Markdown version of this page
License: Use inside your own business and your clients' businesses. Do not resell or republish the bundle itself. Download terms

Related

Prompt packFree

Pursue or Pass: five rules for every federal notice, and a five-check script (free sample)

by PaperChaseWebb

Five rules for a bid/no-bid call you can defend: gate before you score, the score sets a ceiling, write the reason, you decide by name, keep every row. A read-only script checks your list against them in seconds.

Works withClaude CodeCodexGrok BotHermesPlain .md plus one Python 3 script, standard library only; one prompt for all four harnesses

Solo business · Beginner · 5 min setup

Prompt pack$39

GovCon Opportunity Triage Kit: bid or no-bid on a federal opportunity, gates first, scores with evidence

by PaperChaseWebb

Every bid you chase spends hours you could put into one you can win. This kit gives you one repeatable call on every sources sought, RFQ or RFP: eight eligibility gates first (any No stops the bid), then six weighted criteria scored 0 to 5 with the evidence written beside each score, a pursuit cost worksheet with expected value, and four decision bands with overrides. Every rule points to its FAR, 13 CFR or SAM.gov source.

Works withClaude CodeCodexGrok BotHermesPlain Markdown with an agent rules block

Solo business · Intermediate · 15 min setup

Prompt pack$29

Eligibility and Compliance Gate Checklist: eight gates to clear before you bid a set-aside

by PaperChaseWebb

Before you bid a set-aside, check the current SBA and FAR rules on size and status and confirm you are eligible. This is not legal advice. This checklist walks the eight gates in order, from an active SAM record and size under the assigned NAICS to set-aside status, exclusions, the limitations on subcontracting, teaming, responsibility and procurement integrity, with checkboxes and the exact rule behind each one.

Works withClaude CodeCodexGrok BotHermesPlain Markdown with an agent rules block

Solo business · Intermediate · 15 min setup

Prompt pack$24

Evidence-Only Capability Statement: every line traced to proof

by PaperChaseWebb

A one-page federal capability statement your agent can draft without inventing anything. Every competency, past job and status sits in a claim ledger with the file that proves it. A checker traces each line back, refuses invented figures, stale SAM data, uncertified set-aside statuses and praise words, and your signature is tied to the exact text.

Works withClaude CodeCodexGrok BotHermesThree Python scripts, a claim ledger and company record template, a worked example, AGENTS.md and CLAUDE.md rules, a skill file, a Grok Bot standing order and Codex and Hermes notes

Solo business · Beginner · 20 min setup

Prompt pack$19

Government Buyer Outreach Screen: where agencies find vendors and what not to send

by PaperChaseWebb

Commercial outreach habits do not carry over to federal buying. This kit maps seven common places agencies look for small businesses, then screens every planned contact before it goes out: the wrong person during an open solicitation, questions about other bidders, job talk, gifts over the $20 and $50 limits, and any message without your yes on it.

Works withClaude CodeCodexGrok BotHermesA Python screen with tests, a contact log template, four method files, AGENTS.md rules with an owner approval step and two prompts

Solo business · Beginner · 15 min setup

Prompt pack$29

Veteran GovCon AI Kit: capability statement, SAM.gov profile checklist and past performance with your agent

by PaperChaseWebb

For veteran-owned and other small firms starting in federal contracting: a fact sheet your agent drafts from, eight prompts for a one-page capability statement, a SAM.gov profile checklist, a past performance skill, agency research prompts and capstat_check.py, thirteen read-only checks for UEI, CAGE and NAICS shape, hype words, placeholders, length and a dated source line for any status you list.

Works withClaude CodeCodexCursorPlain .md and one Python 3 script, standard library only; tested on Linux with Python 3.13.5, not loaded inside each agent

Solo business · Beginner · 45 min setup