> ## Documentation Index
> Fetch the complete documentation index at: https://docs.coconut.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Competitor Intelligence Loop

> A weekly read on your named competitors, kept fresh as the landscape changes

## What you get

A recurring loop that tracks a named list of competitors and keeps two kinds of documents fresh on every run: one knowledge file per competitor with the current state of their product, pricing, positioning, funding, hiring, and sentiment — and a single landscape overview that summarizes the field at a glance.

The skill picks its run mode automatically:

* **First-run.** No `competitor-{name}.md` yet for this competitor — write a full dated baseline.
* **Delta-run.** File already exists — append a dated changelog with only what materially changed since the last entry.

Good fit when you have a defined competitive set you want to track on a steady cadence, and you'd rather one source per competitor that evolves over time than a fresh research report every week.

## What's inside

* **Skill:** `competitor-intelligence-loop` (rename if you'd like)
* **Job:** Run weekly, or on whatever cadence matches your competitive cycle
* **Knowledge files:**
  * `competitors-list.md` — your editable list of competitors to track (input)
  * `competitor-{name}.md` — one per competitor, named automatically (output)
  * `competitors.md` — landscape overview, rewritten every run (output)

Each competitor file has a dated baseline at the top and a changelog of dated deltas below. The landscape overview reads at a glance — who moved where, what's quiet, what's heating up.

## Set it up

Paste the prompt below into your skill builder. It's standalone — it doesn't depend on any specific platform, only on having a `competitors-list.md` you can edit and a workspace knowledge base the skill can read from and write to.

> I want a skill called `competitor-intelligence-loop` that tracks a named list of competitors on a recurring cadence and synthesizes product, pricing, positioning, funding, hiring, and sentiment signals into per-competitor knowledge files plus a single landscape overview. Each run should distinguish a first-run (full baseline) from a delta-run (only what materially changed) automatically.
>
> **Role.** You are an experienced operator who is skeptical by default. Your job is not to be enthusiastic; it is to produce a concrete, useful output grounded in available context.
>
> **Read first.**
>
> * Any project-level identity or config files in the workspace.
> * Any role, team, or memory notes (`role.md`, `team.md`, `memory.md`, or equivalents).
> * The editable competitor list at `competitors-list.md` in the knowledge base.
> * The landscape overview at `competitors.md` if it exists.
> * Each existing `competitor-{name}.md` file in the knowledge base.
>
> **Before proposing anything, answer these out loud.**
>
> * What is the concrete user or business goal of this run, in one sentence?
> * What evidence in the role notes, memory, knowledge base, or connector data supports doing this now?
> * What would make the output duplicate, shallow, or not worth producing?
>
> **Detect run mode for each competitor.** If `competitor-{name}.md` does not exist yet, this is a first-run for that competitor — write a full dated baseline. If it already exists, this is a delta-run — append only what materially changed since the last entry.
>
> **For each competitor on the list, research and synthesize:**
>
> * **Product and pricing.** Recent releases, pricing changes, new tiers, removed offerings.
> * **Positioning and messaging.** Website hero copy, audience framing, comparison pages.
> * **Funding and hiring.** Funding events, leadership moves, distinctive job openings.
> * **Buyer and analyst sentiment.** Reviews, social commentary, analyst notes — what real users are saying.
>
> Name the source for each signal. Don't repeat what already lives in the existing file unless it has materially changed.
>
> **Write the output.**
>
> * **First-run.** Write a full dated baseline profile to `competitor-{name}.md`, organized by the categories above.
> * **Delta-run.** Append a dated changelog section with only what materially changed. Don't rewrite the baseline.
> * In both cases, rewrite `competitors.md` so the landscape overview reflects the current state of the field at a glance — who moved where, what's quiet, what's heating up.
>
> **Run summary.** At the end of the run, output a concise summary: signals found, evidence used, and any interpretations you considered and rejected (with reasons).
>
> **Escalate when needed.** If any signal warrants human attention (a real strategic shift, not a routine update), file a high-priority task naming the competitor, the signal, the evidence, and the recommended next action.
>
> **Hard bans — do not produce.**
>
> * Generic strategy with no named competitor, artifact, or next action.
> * A task or finding that duplicates an existing one — check the existing task list first.
> * Claims about external data sources you didn't actually check — only cite a source if you read it.
>
> **For every artifact you produce, include:**
>
> * The grounded finding or update.
> * The evidence used (named sources).
> * Any alternative interpretations you considered and why you rejected them.
> * The exact next action — for the operator, the team, or the next run of the skill.
>
> Once we've agreed on the skill, install it and set up a recurring job (weekly is a good starting cadence).

## How the two run modes differ

A **first-run** for a competitor produces a full baseline — every section populated with the current state, dated. A **delta-run** assumes the baseline is correct and appends only what changed since the last entry: a new pricing tier, a leadership departure, a sentiment shift, a positioning move. If nothing material changed, the delta-run says so explicitly rather than padding the file.

On every run — first or delta — the landscape overview at `competitors.md` is rewritten to reflect the current picture across all tracked competitors.

## What the output looks like

A `competitor-{name}.md` file after a first-run plus a couple of deltas:

```markdown theme={null}
# [Competitor] — Intelligence File

**Tracked since:** 2026-04-01
**Last updated:** 2026-05-13

## Baseline (2026-04-01)

### Product and pricing
- ...

### Positioning and messaging
- ...

### Funding and hiring
- ...

### Buyer and analyst sentiment
- ...

## Changelog

### 2026-05-13
- **Pricing.** Introduced a $200/mo team tier (source: pricing page, 2026-05-12). Squeezes the gap between their solo plan and the previous $500 team tier.
- **Hiring.** Two senior PM roles opened in the EU — first time hiring outside the US.
- *Rejected:* "Aggressive expansion into Europe" — a single hiring signal isn't enough to claim a strategic shift; revisit if a leadership hire follows.

### 2026-05-06
- **Sentiment.** Forum threads since the last release lean positive on speed, mixed on reliability.
```

A `competitors.md` landscape overview, rewritten every run:

```markdown theme={null}
# Competitive Landscape

**Last updated:** 2026-05-13
**Tracked competitors:** 4

## At a glance
- **[Competitor A]** — Quiet week. No material moves.
- **[Competitor B]** — New pricing tier squeezing the mid-market.
- **[Competitor C]** — Hiring signal points to EU expansion (watch).
- **[Competitor D]** — Repositioning around enterprise; consumer messaging fading.
```

## Keep going

Edit `competitors-list.md` to add or remove competitors at any time. The next run picks up the change automatically — new names get a first-run baseline; removed names stop receiving deltas, but their file stays put as a historical record.