@opendirectory.dev/skills 0.1.50 → 0.1.52

package/skills/linkedin-job-post-to-buyer-pain-map/evals/evals.json

@@ -0,0 +1,36 @@
+{
+  "skill_name": "linkedin-job-post-to-buyer-pain-map",
+  "evals": [
+    {
+      "id": 1,
+      "prompt": "Analyze this job post for buyer pain. My product is Metabase, an open-source BI tool for non-technical teams. ICP: B2B SaaS, 50-500 employees, Series A-C, modern data stack. Job post: Acme Analytics (Series B, 120 employees) is hiring a Senior Data Platform Engineer to build a self-serve analytics layer for product managers and business leads. Responsibilities include owning ETL and dbt transformations, optimizing query performance for BI tools, and partnering with 6+ internal teams. Requirements: 5+ years with Snowflake, dbt, Airflow.",
+      "expected_output": "Agent checks GEMINI_API_KEY (set). Captures product brief (Metabase, open-source BI), ICP (B2B SaaS, 50-500, Series A-C, modern data stack), and job post. Extracts signals: team=Data Platform, seniority=Senior IC, emphasis=building new (self-serve analytics layer), stack hints=Snowflake, dbt, Airflow. Scores with the LLM: signal_strength 8-9 (explicit analytics pain, modern data stack match), urgency 7-8 (6+ teams partnering, scaling rapidly), icp_fit 8-9 (Series B, 120 employees, dbt+Snowflake). Overall score 80-90. Buy-vs-build: Hybrid (buy-and-build). Stage: Scaling post-PMF. Primary pain: data team bottleneck for business teams needing metrics access. Evidence cites 'self-serve analytics layer' and 'without pinging the data team'. Outreach angle references self-serve analytics build-out. Handoff block includes suggested personas (Head of Data, VP Product). Saves to docs/pain-maps/acme-analytics.md.",
+      "files": []
+    },
+    {
+      "id": 2,
+      "prompt": "Decode this job description for GTM signals. Product: Ramp, a sales onboarding platform. ICP: B2B SaaS, 100-1000 employees, 10+ sales reps, Series B+. Job post: CoolStartup Inc is hiring a Full Stack Engineer. 'We're looking for a rockstar full-stack engineer who loves building cool things. React, Node.js, PostgreSQL. Must be a team player in a fast-paced startup. We offer equity, unlimited PTO, and weekly team lunches.'",
+      "expected_output": "Agent checks GEMINI_API_KEY (set). Captures inputs. Extracts signals: team=Engineering, seniority=Mid-level IC, emphasis=generic (no clear pain), stack hints=React/Node/PostgreSQL (unrelated to sales onboarding). Scores: signal_strength 1-2 (no sales, onboarding, or GTM signals), urgency 1-2 (no urgency language, exploratory tone), icp_fit 1-2 (no company size, no sales team indicators). Overall score 10-20. Buy-vs-build: Unknown. Stage: Unknown. No primary pains inferred — agent explicitly states post is too generic to extract meaningful pain signals. Does not hallucinate pains. Recommends skipping this account. No outreach angles generated because signal strength is too low.",
+      "files": []
+    },
+    {
+      "id": 3,
+      "prompt": "Build a pain map from this hiring post. Product: Datadog, monitoring and observability for cloud-native apps. ICP: cloud-native tech companies, 200+ engineers, Kubernetes, multi-cloud, Series C+ or public. Job post: MedSecure Health Systems (2,000 employees, healthcare technology) is hiring a Director of Cloud Infrastructure to lead cloud migration of 200+ on-premise applications to AWS, build container orchestration capabilities, and establish monitoring and observability practices from scratch. Requirements: AWS, Terraform, Kubernetes, HIPAA compliance.",
+      "expected_output": "Agent checks GEMINI_API_KEY (set). Captures inputs. Extracts signals: team=IT Infrastructure, seniority=Director/Leadership, emphasis=building new (from scratch) and replacing (migrating on-prem), stack hints=AWS/Terraform/Kubernetes. Scores: signal_strength 7-9 (Director-level, building observability from scratch, Kubernetes), urgency 7-8 (200+ app migration, leadership hire), icp_fit 2-4 (healthcare not cloud-native tech, on-prem migrating not cloud-native). Overall score 45-60. Stage: Mature/Enterprise. Buy-vs-build: Leaning buy. Primary pain: no observability stack exists, migrating 200+ apps blind. Evidence cites 'establish monitoring and observability practices from scratch'. Agent flags account as OUTSIDE CORE ICP with recommendation to deprioritize unless healthcare is an expansion segment. Still provides outreach angle in case user wants to pursue.",
+      "files": []
+    },
+    {
+      "id": 4,
+      "prompt": "Analyze these job posts for buyer pain signals.",
+      "expected_output": "Agent detects GEMINI_API_KEY is missing from environment. Stops at Step 1. Tells the user: 'GEMINI_API_KEY is required. Get it at aistudio.google.com. Add it to your .env file.' Does not proceed to collect inputs or generate any analysis.",
+      "files": [],
+      "setup": "GEMINI_API_KEY not set in environment"
+    },
+    {
+      "id": 5,
+      "prompt": "Analyze these 3 job posts from the same company. Product: Gong, revenue intelligence platform. ICP: B2B SaaS, 200+ employees, 20+ sales reps, Series B+. Company: RapidScale (Series C, 450 employees). Post 1: VP of Sales — build and scale the outbound sales org from 15 to 50 reps over 18 months, establish sales methodology and playbooks, own pipeline generation strategy. Post 2: Revenue Operations Manager — implement and manage CRM workflows, build forecasting models, reduce manual pipeline reporting, first RevOps hire. Post 3: Senior Customer Success Manager — manage 40+ enterprise accounts, reduce churn in enterprise segment, build QBR process from scratch.",
+      "expected_output": "Agent checks GEMINI_API_KEY (set). Captures inputs. Groups all 3 posts under RapidScale. Extracts signals: Post 1 (GTM/Sales leadership, scaling 15→50 reps, building playbooks), Post 2 (RevOps, first hire, CRM + forecasting, manual pipeline reporting), Post 3 (CS, enterprise churn, QBR process). Identifies dominant theme: GTM infrastructure buildout across Sales + RevOps + CS. Scores: signal_strength 9-10 (multiple related hires, explicit GTM pain, VP-level), urgency 8-9 (scaling 15→50, first RevOps hire, enterprise churn), icp_fit 8-9 (Series C, 450 employees, 20+ reps). Overall score 85-95. Stage: Scaling post-PMF. Buy-vs-build: Hybrid (buy-and-build). Primary pains: (1) No sales methodology or playbooks for scaling team, (2) Manual pipeline reporting with no RevOps function, (3) Enterprise churn with no structured QBR process. Multiple outreach angles generated. Handoff block targets VP Sales and RevOps Manager as personas. Report groups all 3 signals under one company entry.",
+      "files": []
+    }
+  ]
+}

package/skills/linkedin-job-post-to-buyer-pain-map/references/examples.md

@@ -0,0 +1,122 @@
+# Worked Examples
+
+This file provides 3 worked examples showing how the skill interprets different types of job posts. The agent reads this file to calibrate its analysis quality.
+
+---
+
+## Example 1: High Signal, High ICP Fit
+
+### Input
+
+**Product brief:** "Metabase — open-source BI tool that lets non-technical teams query data and build dashboards without SQL. Target: Series A–C startups with growing data teams."
+
+**ICP description:** "B2B SaaS companies, 50–500 employees, Series A–C, have at least 2 data engineers, use modern data stack (dbt, Snowflake, BigQuery), need self-serve analytics for business teams."
+
+**Job post:**
+```json
+{
+  "company_name": "Acme Analytics",
+  "job_title": "Senior Data Platform Engineer",
+  "location": "Remote, US",
+  "seniority": "Senior / IC",
+  "team_or_function": "Data Platform / Infra",
+  "job_description_text": "We're looking for a Senior Data Platform Engineer to own our analytics infrastructure. You'll build a self-serve analytics layer so product managers and business leads can access metrics without pinging the data team. Responsibilities: own ETL pipelines and dbt transformations, optimize query performance for our BI tools, partner with 6+ internal teams to define metrics, and ensure data reliability for executive dashboards. Requirements: 5+ years with modern data stack (Snowflake, dbt, Airflow), experience building self-serve data products, strong SQL. We are a Series B company with 120 employees scaling rapidly."
+}
+```
+
+### Expected Output
+
+**Summary:**
+- Headline: "Staffing data infra to unblock self-serve analytics for non-technical teams"
+- Overall score: 87
+- Signal strength: 9 (explicit analytics pain + modern data stack + building self-serve layer)
+- Urgency: 8 (partnering with 6+ teams suggests backlog, "scaling rapidly")
+- ICP fit: 9 (Series B, 120 employees, dbt + Snowflake, self-serve analytics need)
+- Stage guess: Scaling post-PMF
+- Buy-vs-build: Hybrid (buy-and-build)
+  - _Reasoning: building infrastructure + likely to buy BI tooling_
+
+**Primary pain:** "Data team is a bottleneck for business teams needing metrics access. PMs and leads are blocked by ad-hoc data requests."
+
+**Evidence cited:** "product managers and business leads can access metrics without pinging the data team", "partner with 6+ internal teams"
+
+**Outreach angle:** "Reduce ad-hoc data fire drills — lead with how Metabase gives PMs self-serve dashboards so the data team stops being a ticket queue."
+
+---
+
+## Example 2: Low Signal, Generic Post
+
+### Input
+
+**Product brief:** "Ramp — sales onboarding platform for B2B SaaS. Helps VP Sales cut new-rep ramp time."
+
+**ICP description:** "B2B SaaS, 100–1000 employees, at least 10 sales reps, hiring SDRs/AEs, series B+."
+
+**Job post:**
+```json
+{
+  "company_name": "CoolStartup Inc",
+  "job_title": "Full Stack Engineer",
+  "location": "San Francisco, CA",
+  "seniority": "Mid-level / IC",
+  "team_or_function": "Engineering",
+  "job_description_text": "We're looking for a rockstar full-stack engineer who loves building cool things. You'll work on our product using React, Node.js, and PostgreSQL. Must be a team player who thrives in a fast-paced startup environment. We offer competitive salary, equity, unlimited PTO, and weekly team lunches. Come join our amazing culture!"
+}
+```
+
+### Expected Output
+
+**Summary:**
+- Headline: "Generic engineering hire with no relevant GTM or sales signals"
+- Overall score: 16
+- Signal strength: 1 (no sales, onboarding, or GTM signals; engineering role unrelated to product area)
+- Urgency: 2 (no urgency language, exploratory tone)
+- ICP fit: 2 (no company size, no sales team indicators, no relevant function)
+- Stage guess: Unknown (no indicators)
+- Buy-vs-build: Unknown
+
+**Primary pain:** None identified. Post is too generic to infer operational pain relevant to sales onboarding.
+
+**Note to user:** "This post contains minimal signal for your ICP. The role is a general engineering hire with no references to sales operations, onboarding, or GTM infrastructure. Recommend skipping."
+
+---
+
+## Example 3: Decent Pain, Out-of-ICP
+
+### Input
+
+**Product brief:** "Datadog — monitoring and observability platform for cloud-native applications. Target: tech companies running Kubernetes in production."
+
+**ICP description:** "Cloud-native tech companies, 200+ engineers, running Kubernetes/containers, multi-cloud, Series C+ or public."
+
+**Job post:**
+```json
+{
+  "company_name": "MedSecure Health Systems",
+  "job_title": "Director of Cloud Infrastructure",
+  "location": "Chicago, IL",
+  "seniority": "Director / Leadership",
+  "team_or_function": "IT Infrastructure",
+  "job_description_text": "MedSecure Health Systems is seeking a Director of Cloud Infrastructure to lead our cloud migration initiative. You will oversee the transition of 200+ on-premise applications to AWS, build container orchestration capabilities, and establish monitoring and observability practices from scratch. Manage a team of 8 infrastructure engineers. Requirements: 10+ years in infrastructure, experience with AWS, Terraform, container orchestration (Kubernetes preferred), HIPAA compliance expertise required. We are a 2,000-employee healthcare technology company."
+}
+```
+
+### Expected Output
+
+**Summary:**
+- Headline: "Healthcare company building cloud infra and observability from scratch"
+- Overall score: 62
+- Signal strength: 8 (Director-level, building observability from scratch, Kubernetes, AWS migration)
+- Urgency: 7 (200+ app migration, building from scratch, leadership hire)
+- ICP fit: 3 (healthcare, not cloud-native tech company; on-prem migrating to cloud, not cloud-native)
+- Stage guess: Mature / Enterprise
+- Buy-vs-build: Leaning buy
+  - _Reasoning: "establish monitoring and observability practices from scratch" suggests they need to adopt tooling_
+
+**Primary pain:** "No monitoring or observability stack exists. Migrating 200+ on-prem apps to AWS with zero cloud-native visibility."
+
+**Evidence cited:** "establish monitoring and observability practices from scratch", "transition of 200+ on-premise applications to AWS"
+
+**ICP flag:** ⚠️ Outside core ICP — healthcare vertical, on-prem migrating (not cloud-native). Pain is real and product-relevant, but company profile does not match ideal customer profile. Consider deprioritizing unless healthcare is an expansion segment.
+
+**Outreach angle:** "If pursued: lead with migration observability — reference the 200-app migration and offer a proof-of-value on the first 10 apps moved to AWS."

package/skills/linkedin-job-post-to-buyer-pain-map/references/scoring-rubric.md

@@ -0,0 +1,120 @@
+# Scoring Rubric
+
+This file defines the scoring model used by the linkedin-job-post-to-buyer-pain-map skill. The agent reads this file during Step 4 to evaluate each account.
+
+---
+
+## Three Scoring Dimensions
+
+### Signal Strength (1–10)
+
+Measures how many and how specific the hiring signals are that relate to the user's product area.
+
+**Scoring rules:**
+
+| Condition | Modifier |
+|-----------|----------|
+| Multiple related hires on the same team or problem area | +3 |
+| Explicit language about a pain area the user's product addresses | +2 |
+| Senior role (Head / Director / VP) in a relevant function | +1 |
+| Role responsibilities reference building, replacing, or fixing a system in the product's category | +1 |
+| "First X hire" or "0→1" language indicating new team formation | +1 |
+| Generic "nice-to-have" requirements with no clear pain signal | −1 |
+| Job post is mostly a list of perks and culture statements with minimal responsibilities | −2 |
+
+**Floor:** 1. **Ceiling:** 10. Clamp after summing.
+
+**Base:** Start at 3 for any post that mentions a function relevant to the user's ICP.
+
+---
+
+### Urgency (1–10)
+
+Measures how time-sensitive the hiring need appears from the text.
+
+**Scoring rules:**
+
+| Condition | Modifier |
+|-----------|----------|
+| Explicit urgency language: "immediate", "ASAP", "critical hire", "backfill" | +3 |
+| Multiple open roles in the same function at the same company | +2 |
+| Language suggesting firefighting: "stabilize", "fix", "unblock", "reduce downtime" | +2 |
+| Role described as replacement for a departing leader | +1 |
+| "Nice to have" tone, exploratory language ("we're thinking about...") | −2 |
+| No urgency indicators at all | −1 |
+
+**Floor:** 1. **Ceiling:** 10. Clamp after summing.
+
+**Base:** Start at 4 for any role at IC-Senior or higher that has a defined scope.
+
+---
+
+### ICP Fit (1–10)
+
+Measures how closely the company profile in the job post matches the user's `icp_description`.
+
+**Scoring rules:**
+
+| Condition | Modifier |
+|-----------|----------|
+| Industry matches ICP exactly | +3 |
+| Company size indicators match (headcount, "Series X", "enterprise") | +2 |
+| Tech stack or tool references overlap with ICP tech stack | +2 |
+| Role function matches ICP target persona function | +1 |
+| Industry is adjacent but not exact match | +1 |
+| Industry is clearly outside ICP (e.g., government when ICP is B2B SaaS) | −3 |
+| Company size indicators suggest a segment the user does not serve | −2 |
+
+**Floor:** 1. **Ceiling:** 10. Clamp after summing.
+
+**Base:** Start at 3 for any post where industry cannot be determined.
+
+---
+
+## Overall Score (10–100)
+
+```
+overall = round((0.4 × signal_strength + 0.3 × urgency + 0.3 × icp_fit) × 10)
+```
+
+Each dimension is 1–10, so the raw weighted sum ranges from 1.0 to 10.0. Multiply by 10 to get 10–100 scale.
+
+**Interpretation guide:**
+
+| Range | Label | Recommendation |
+|-------|-------|---------------|
+| 80–100 | Strong signal | Prioritize immediately. Build pain map and outreach angles. |
+| 60–79 | Moderate signal | Worth analyzing. Pain map likely has actionable pains. |
+| 40–59 | Weak signal | Low priority. Limited pain signal. |
+| 1–39 | Noise | Skip or deprioritize. Post is too generic or outside ICP. |
+
+---
+
+## Buy-vs-Build Inference
+
+The skill infers the company's likely posture toward buying vs building based on language patterns in the job description.
+
+| Language Pattern | Inferred Posture |
+|-----------------|-----------------|
+| "Build from scratch", "0→1", "greenfield", "design and implement new" | **Leaning build** |
+| "Evaluate and implement", "select vendor", "integrate third-party" | **Leaning buy** |
+| "Replace legacy", "migrate from", "modernize" | **Leaning buy** (replacing existing) |
+| "Own end-to-end", "build and maintain" + "integrate tools" | **Hybrid (buy-and-build)** |
+| "Scale existing", "optimize current" | **Hybrid** — already built, may buy to augment |
+| No clear signals | **Unknown** — do not guess |
+
+Output as a text label, not a numeric score.
+
+---
+
+## Stage Guess
+
+Inferred from company context clues in the job description.
+
+| Clue | Stage Guess |
+|------|------------|
+| "Series A", "seed", "early stage", < 50 employees | Pre-PMF or Early |
+| "Series B/C", "scaling", "hypergrowth", 50–500 employees | Scaling post-PMF |
+| "Enterprise", "Fortune 500", "publicly traded", > 500 employees | Mature / Enterprise |
+| "First [role] hire", "founding team" | Very early |
+| No clues available | "Unknown" |