Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Doris-Labs/sales-skills --skill crm-hygienegit clone --depth 1 https://github.com/Doris-Labs/sales-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/doris-labs/sales-skills/crm-hygiene)<a href="https://agentmods.dev/skills/doris-labs/sales-skills/crm-hygiene"><img src="https://agentmods.dev/badge/skills/doris-labs/sales-skills/crm-hygiene/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/doris-labs/sales-skills/crm-hygiene"><img src="https://agentmods.dev/badge/skills/doris-labs/sales-skills/crm-hygiene.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00067 | $0.01388 |
| Opus 5 | $0.00034 | $0.00694 |
| Sonnet 5 | $0.00013 | $0.00278 |
| Haiku 4.5 | $0.00007 | $0.00139 |
Grade A, and why
crm-hygiene scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CRM Hygiene
Purpose
Take a set of CRM deals and find where the record has drifted from reality — stale fields, impossible dates, stages that outrun the evidence — then hand back concrete, approve-able corrections so the pipeline you forecast on is the pipeline you actually have.
Inputs
- The deals to check (a single deal, an owner's open pipeline, or one stage)
- Current CRM field values: stage, amount, close date, next step, owner, last activity
- Evidence of what's really happening: recent meetings, commitments, stakeholders
Method
Run three passes. Each produces flags; the output is a ranked list of write-back proposals.
1. Field-by-field staleness check. For every deal, test each field against a rule:
- Next step — is one set, and is it dated and in the future? A blank or past-dated next step means the deal has no defined motion. Flag.
- Close date — is it in the past? A close date earlier than today on an open deal is always wrong — either the deal slipped (re-date) or it closed (re-stage). Flag.
- Stage — does it match the last real interaction? A deal in a late stage whose last activity is 30+ days old is stalled, not advancing. Flag.
- Amount — is it set and non-zero past early stages? An unvalued deal in mid/late stage can't be forecast. Flag.
- Owner / last activity — is there activity in the last 14 days on an open deal? No touch = at risk of going dark. Flag.
2. Contradiction detection. Cross-field logic the AE knows but the CRM doesn't enforce:
- Stage is late (negotiation/proposal) but no economic buyer / decision-maker is on the deal → stage is ahead of access.
- Stage is late but no recent activity → forecast-padding, not a live deal.
- Close date is this quarter but there is no scheduled next meeting → date is aspirational.
- Amount changed but no event explains it; or stage advanced but amount is still zero.
3. Stage-vs-evidence mismatch. Hold the CRM stage up against what the conversations actually contain. Rule of thumb — a stage is only earned when its proof exists:
- Discovery needs an identified pain.
- Qualified needs a confirmed metric + economic buyer.
- Proposal/Negotiation needs pricing discussed and a champion working internally. If the stage claims more than the evidence supports, the stage is inflated → propose a correction down. If the evidence supports more than the stage shows, propose advancing.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 104 lines · 67 tokens per session scan A ff3d733cc658
crm-hygiene is a skill published in the GitHub repository Doris-Labs/sales-skills (3 stars, last pushed 3mo ago), licensed MIT. It adds 67 tokens to every session and 1,388 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
lynqu
Entry point for the Lynqu sales suite — describe any sales situation and it composes the right skills across capture, outreach, quoting and reporting. Requires the Lynqu MCP server connected.
lynqu-sales-playbook
Turn a written sales playbook into a live Lynqu pipeline — contacts, timed tasks, follow-up templates, the playbook attached to each lead, and stage rules that deliver the prep at the right moment.
lynqu-card-studio
Create and update Lynqu digital business cards — contact info, social links, services, template, palette — and check card views, scans and shares. Requires the Lynqu MCP server connected.
lynqu-competitors
Build a battlecard for the incumbent on a Lynqu deal — switching cost, wedge, objections and honest answers — written back to the lead. Requires the Lynqu MCP server.
lynqu-contacts
Map the buying committee on an account — economic buyer, champion, blocker — and write it into Lynqu as contacts and participants. Requires the Lynqu MCP server.
lynqu-deal-desk
Price a Lynqu deal and draft the quote — browse the price book, put priced lines on a deal, then produce a draft quote for a human to send. Requires the Lynqu MCP server connected.