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 Frontal-so/outbound-skills --skill debugginggit clone --depth 1 https://github.com/Frontal-so/outbound-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/frontal-so/outbound-skills/debugging)<a href="https://agentmods.dev/skills/frontal-so/outbound-skills/debugging"><img src="https://agentmods.dev/badge/skills/frontal-so/outbound-skills/debugging/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/frontal-so/outbound-skills/debugging"><img src="https://agentmods.dev/badge/skills/frontal-so/outbound-skills/debugging.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 | $0.00124 | $0.01281 |
| Opus 5.5 | $0.00050 | $0.00512 |
| Sonnet 5.5 | $0.00025 | $0.00256 |
| Haiku 4.5 | $0.00012 | $0.00128 |
Grade A, and why
debugging 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 1mo 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 — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Debugging Clay Workflows
You help users diagnose and fix problems in their Clay tables, enrichments, and automations.
References
- Read
{SKILL_BASE}/resources/expert-tips-eric-noski.mdfor the 6 table rules and 8 AI prompting rules. - Read
{SKILL_BASE}/resources/workflow-patterns.mdfor auto-update behavior and run settings.
Top 10 Common Mistakes
1. No Conditional Runs on Paid Enrichments
Symptom: Credits burning on rows that already have data.
Fix: Add conditional run to every paid column: /field is empty. Non-negotiable.
2. Auto-Update Enabled During Testing
Symptom: Credits consumed every time new rows are added. Fix: Set table-level auto-update to OFF while building. Only enable when workflow is finalized.
3. Wrong Input Column Mapped
Symptom: Enrichment returns empty or wrong results. Fix: Verify input mapping. Email waterfall needs domain, not company name. LinkedIn enrichment needs profile URL, not company URL.
4. AI Prompt Doing Too Many Things
Symptom: Claygent/AI returns inconsistent or partial results. Fix: One task per AI column. Split "check site + classify + extract" into 3 separate columns. GPT-4 Mini for each.
5. No Null Handling in Formulas
Symptom: Formula errors, "undefined" in cells.
Fix: Add || "" to every variable reference: ({{field}} || ""). Always handle empty cells.
6. Duplicate Rows Consuming Double Credits
Symptom: Same contact enriched multiple times. Fix: Enable auto-dedupe before importing. Key: email or LinkedIn URL. Note: case-sensitive ("Clay" != "clay").
7. Using Expensive AI Models for Simple Tasks
Symptom: 3x credit cost with no quality improvement. Fix: GPT-4 Mini / Claygent Neon for 90% of tasks. Only use GPT-4/Claude Opus for complex reasoning.
8. Not Saving Enriched Data Externally
Symptom: Paying to re-enrich the same records in a new table. Fix: Push all enriched data to CRM or Supabase ($30/month for 11.4M+ records). Query before re-enriching.
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.
- 1mo ago First seen · 96 lines · 124 tokens per session scan A 36e90ba1ab87
debugging is a skill published in the GitHub repository Frontal-so/outbound-skills (8 stars, last pushed 29d ago), licensed MIT. It adds 124 tokens to every session and 1,281 once invoked, about $0.0005 per session on Opus 5.5. 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
audience-icp-filter
Filter an existing audience or lead list against your ICP and split it into ready-to-sequence segments. Use when someone already has a list of people — an audience in their sales tool, a CSV or CRM export, event or webinar attendees, registrants, a Sales Navigator import, a newsletter or community export — and wants…
sales-nav-search-builder
Generate a precise LinkedIn Sales Navigator search URL from a natural-language ICP description — for sales prospecting, outbound outreach, lead generation, and B2B targeting. Use whenever the user asks for a Sales Nav URL, a LinkedIn search URL, an outbound prospecting query, a search for a persona or role, wants to…
team-performance-dashboard
Generate your Team Performance Dashboard from your La Growth Machine data: a head-of-sales cockpit that ranks each sender (rep) on reply rate, your success event (signups/meetings/deals) and conversion, surfaces hot leads going cold, and turns your best reps' campaigns into a coaching playbook for the rest. Pulls only…
weekly-performance-advisor
Generate your Weekly Performance Advisor dashboard from your La Growth Machine data: a two-tab cockpit (To do + Weekly performance) that flags campaigns to fix, sorts replies by urgency, and tracks reply volume week over week. It pulls only YOUR live LGM data, scores each running campaign against baked 3-zone…
outreach-icp-finder
Find your PROVEN ideal customer profile from your own outreach data — who actually replies, accepts and shows interest — then find more like them. Reads La Growth Machine via its MCP, or a CSV export from any outreach tool (lemlist, Instantly, Smartlead, HeyReach, Apollo, Waalaxy…). Use when someone asks which job…
objection-analyzer
Find out which objections your outbound gets, how well your team handled them, and what to say next time. Use whenever the user wants to analyze the objections in their replies, rank the most frequent ones, know how to handle a specific objection, coach reps on their objection answers, build an objection playbook or…