Systems Lab

Agent skill

revenue-forecasting

Forecast 30/60/90-day revenue using weighted-pipeline math, historical stage-conversion rates, and conservative/base/aggressive scenarios with explicit assumption sets.

slowingNeeds a keyActs undeclared424 words

Filed under CRM and RevOps.

From devangk003/gtm-agent-skills · 32 skills · 0 · pushed 2026-06-18

What it does when it runs

Forecast 30/60/90-day revenue using weighted-pipeline math, historical stage-conversion rates, and conservative/base/aggressive scenarios with explicit assumption sets. Use when the user says "forecast revenue", "board update numbers", "commit number", or "recalibrate conversion rates.

Read from the skill and the 4 files bundled beside it. A skill’s own description is written to be selected by an agent, so it describes the job and not the dependencies.

Keys and connectors you must supply
  • AGENTIC_APP_TOKEN
Hosts it reaches
No third-party host appears in the skill or its bundled files.
Tool permissions it declares
No allowed-tools in the frontmatter. It does act, so it runs under whatever permissions your session already grants.
Actions present in the files
shell

Ask about revenue-forecasting

Opens your assistant with this page's verified links already in the prompt.

Is this safe to install?ClaudeChatGPT
Adapt it to my stackClaudeChatGPT
What else do I need for it to workClaudeChatGPT
Rather ask a human? Talk to Cheetah

Install it

View source on GitHub ↗
git clone --depth 1 --filter=blob:none --sparse https://github.com/devangk003/gtm-agent-skills.git /tmp/gtm-agent-skills
git -C /tmp/gtm-agent-skills sparse-checkout set "revenue-forecasting"
mkdir -p ~/.claude/skills/revenue-forecasting
cp -R "/tmp/gtm-agent-skills/revenue-forecasting/." ~/.claude/skills/revenue-forecasting/

Picked up without a restart. A project skill of the same name is shadowed by your personal one. For one repository only, swap ~/.claude/skills for .claude/skills. Claude Code docs ↗

The folder is the same in every client that implements the format — 46 of them — so if yours is not above, only the destination changes.

Before you install: this skill will not complete its job on a bare agent. It needs AGENTIC_APP_TOKEN, which you have to obtain separately.

The skill

Source on GitHub ↗

Reproduced in full from devangk003/gtm-agent-skills/blob/6b9a1b0094ffd83d6c02bc03b7ce1610661a1644/revenue-forecasting/SKILL.md, which is licensed MIT (skill frontmatter). 424 words, 6 headings.

Revenue Forecasting

Produce 30/60/90-day revenue forecasts using weighted-pipeline math + historical stage-conversion rates + 3 scenarios. Each forecast carries an explicit assumption set so the user can challenge inputs, not just outputs. Hard rule: forecasts without ≥30 historical closes carry a confidence: low flag.

When to Use

  • Board update — produce 30/60/90 forecast with scenarios
  • Quarter-end commit number
  • Stage-conversion rates need recalibrating
  • Forecast accuracy audit vs actuals
  • User says "forecast revenue" or "what's our commit number"

Quick Reference

ConceptValue
Base scenarioΣ(deal_value × stage_probability) for in-horizon deals
ConservativeProposal-stage + explicit commits only
AggressiveConservative + best-case ≥Discovery (cap +15pp)
Default stage probabilitiesNew 2% / Contacted 5% / Engaged 10% / Meeting 20% / Discovery 35% / Proposal 60%
Confidence rubricHigh: ≥30 closes / Medium: 10–29 / Low: 1–9 / Hypothesis-only: 0
Cycle-time gatingDon't include deals that can't realistically close in horizon

Procedure

  1. Validate inputs. Pull active pipeline; load historical conversion rates (or flag use-of-defaults); confirm horizon.
  2. Filter to in-horizon deals. Deals with expected_close_date within horizon. Exclude implausible stage-close combos. See ${HERMES_SKILL_DIR}/references/scenario-math.md.
  3. Compute base scenario. For each in-horizon deal: contribution = deal_value × stage_probability. Sum.
  4. Compute conservative. Proposal-stage deals at historical rate + explicit commits at 90%.
  5. Compute aggressive. Conservative + best-case ≥Discovery. Cap each deal's added probability at base+15pp.
  6. Per-deal contribution + top-10. List top-10 deals by weighted contribution. Surface concentration risk if >70%.
  7. Output assumption set. Stage probabilities used, cycle-time averages, confidence tier, source of rates. See ${HERMES_SKILL_DIR}/references/assumption-set.md.
  8. Forecast-accuracy audit (when applicable). Compare prior forecasts to actuals. Compute MAPE per scenario. Flag systemic over/under-forecasting.
  9. Push to CRM. Forecast as interaction:research with full assumption set + per-scenario numbers. See ${HERMES_SKILL_DIR}/scripts/push_to_crm.py.

Pitfalls

  • Treating base as THE forecast — it's the middle of three; commit at conservative
  • Using defaults when history exists — always calibrate from actuals
  • Cycle-time blind spots — New-stage deal closing in 7d is implausible
  • Top-10 dependency hidden — surface concentration risk
  • Open rate in forecast — Apple MPP made it noise; stage-based math only
  • Aggressive with no cap — limit upside inflation at +15pp per deal

Verification

  1. Per-scenario math reproducible from assumption set
  2. Top-10 contribution sums plausibly to scenario totals
  3. Confidence tier matches historical-close evidence
  4. Cycle-time exclusions documented
  5. Assumption set explicit (no hidden defaults)

Files bundled with it

These load only when the skill asks for them, so they cost nothing until it runs.

Other skills for the same job

Different authors, same problem. Matched on the words in the skill name, across every library in the catalogue except this one.

Need help setting it up?

This page tells you what revenue-forecasting does and what it needs. Cheetah builds the agent setup it runs inside: data, CRM, sequencing and the guardrails.

Book a call →

The directory stays free. There is nothing gated behind this.