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 andrew-shwetzer/career-ops-plugin-do-not-fork-currently-updating-v2- --skill outreachgit clone --depth 1 https://github.com/andrew-shwetzer/career-ops-plugin-do-not-fork-currently-updating-v2-Wrote 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/andrew-shwetzer/career-ops-plugin-do-not-fork-currently-updating-v2-/outreach)<a href="https://agentmods.dev/skills/andrew-shwetzer/career-ops-plugin-do-not-fork-currently-updating-v2-/outreach"><img src="https://agentmods.dev/badge/skills/andrew-shwetzer/career-ops-plugin-do-not-fork-currently-updating-v2-/outreach/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/andrew-shwetzer/career-ops-plugin-do-not-fork-currently-updating-v2-/outreach"><img src="https://agentmods.dev/badge/skills/andrew-shwetzer/career-ops-plugin-do-not-fork-currently-updating-v2-/outreach.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Prompt Injection · line 105 Subtle instructions detected that may alter agent decision-making or introduce hidden biases.Fix: Review content for implicit steering or bias. Ensure instructions are explicit and align with the skill's stated purpose.
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 | $0.00064 | $0.00901 |
| Opus 5 | $0.00032 | $0.00451 |
| Sonnet 5 | $0.00013 | $0.00180 |
| Haiku 4.5 | $0.00006 | $0.00090 |
Grade B, and why
outreach scanned grade B with 1 finding 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 13d 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.
Subtle steeringmediumPrompt injection
Instructions that bias recommendations or shape behaviour without the user noticing.
- NEVER suggest the user claim a connection that doesn't exist How it starts
The opening of the file, as written. The whole thing — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Draft Outreach
Create personalized outreach messages for job search networking.
Step 0: Load Context
- Read
data/profile.yml - Check
data/research/{company}.mdfor company intelligence - Check
data/evaluations/for any evaluation at this company
If no company research exists:
"I don't have research on {company} yet. Better outreach comes from better intel. Want me to research them first, or should I draft something with what I know?"
Step 1: Identify the Contact
Parse user input for: contact name, title, company, platform.
If no specific contact named, suggest based on research file:
"Based on my research, here are contacts at {company}: {list from research} Who would you like to reach out to?"
Step 2: Determine Message Type
| Type | When | Length |
|---|---|---|
| LinkedIn connection request | No existing connection | Under 300 characters |
| LinkedIn message | Already connected | 100-200 words |
| Cold email | Have their email | 100-150 words |
| Follow-up | Already reached out, no response 5+ days | 50-75 words |
Ask the user which type if not clear from context.
Step 3: Generate Using 3-Part Structure
Part 1: Hook (about THEM, not you)
Reference something specific about the company, their work, or a recent event.
Bad: "I'm really interested in your company" (about you) Bad: "I'd love to connect" (generic) Good: "Your team's work on {specific project/launch} caught my attention" Good: "I noticed {company} just {recent event from research}"
Part 2: Proof (one quantifiable thing about you)
One sentence. One number. Directly relevant to their world.
Bad: "I have 10 years of experience in marketing" Good: "I grew organic traffic 3x at {Company} in 8 months" Good: "I managed a $2M portfolio with 98% client retention"
Pull the most relevant proof point from the user's profile that connects to the target company's needs.
Part 3: Proposal (low-pressure ask)
Bad: "Can you refer me?" (presumptuous) Bad: "I'd love to pick your brain" (vague, one-sided) Good: "Would you be open to a 15-minute chat about what {team} looks for?" Good: "I'd appreciate any advice on standing out for the {role} opening"
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.
- 13d ago First seen · 112 lines · 64 tokens per session scan B fdf402c81888
outreach is a skill published in the GitHub repository andrew-shwetzer/career-ops-plugin-do-not-fork-currently-updating-v2- (490 stars, last pushed 1mo ago), licensed MIT. It adds 64 tokens to every session and 901 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 1 finding (subtle steering). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
hr-tech-expert
Build comprehensive HR technology solutions including HRIS implementation, recruitment automation, performance management, and employee analytics. Use when the user mentions HRIS, an ATS or recruitment pipeline, payroll, onboarding, performance reviews, or employee engagement analytics.
resume-jd-optimizer-cn
A Chinese résumé-editing workflow that matches a résumé to a job description using only confirmed experience and evidence.
resume-tailor
Tailors a resume to a specific job description with ATS optimization. Use when asked to optimize a resume, tailor CV for a job, create an ATS-friendly resume, generate a cover letter, or adapt career profile for a specific role. Triggers on: "resume", "CV", "tailor", "ATS", "job application", "cover letter", "optimize…
ats-resume-formatting
Use this skill whenever a resume needs to be formatted, reviewed, or debugged for upload to Workday, Greenhouse, Lever, iCIMS, or any other ATS (applicant tracking system). Trigger on requests like "format my resume for ATS," "is my resume ATS-friendly," "why is Workday leaving fields blank," "clean up my resume for…
resume-studio
Build, tailor, and maintain job-winning resumes with a persistent career memory. Use this skill WHENEVER the user mentions a resume or CV in any capacity — creating one, updating one, tailoring one for a job description, reviewing/critiquing one, converting a LinkedIn profile into a resume, preparing a job…
rp-help
You are a helpful guide for the Recruiter Skills Pack. Your job is to show the recruiter what tools are available, which ones they can use right now, and which skill to run first based on where they are in their workflow.