feat(skills): add dynamic skill fetching to all router skills (#19)

When an agent matches a skill via a router but doesn't have it installed
locally, the router now instructs it to fetch the SKILL.md directly from
the GitHub repo using the raw content URL pattern:

  https://raw.githubusercontent.com/getsentry/sentry-for-ai/main/skills/<name>/SKILL.md

This ensures agents that only have the three router skills installed can
still access the full library of SDK, workflow, and feature skills
on-demand without requiring a full local clone. The Skill Tree is
referenced as the source of truth for valid skill names and paths.

Updated all three routers:
- sentry-sdk-setup (SDK setup wizard routing)
- sentry-workflow (issue fixing, code review routing)
- sentry-feature-setup (AI monitoring, OTel, alerts routing)

Co-Authored-By: Claude (Anthropic)

Author: HazAT

skills/sentry-feature-setup/SKILL.md

@@ -22,3 +22,12 @@ Match the user's request to the appropriate feature skill below.
 1. If the user mentions AI monitoring, LLM tracing, or instrumenting an AI SDK → load `sentry-setup-ai-monitoring`.
 2. If the user mentions OpenTelemetry, OTel Collector, or multi-service telemetry routing → load `sentry-otel-exporter-setup`.
 3. If the user mentions alerts, notifications, on-call, Slack/PagerDuty integration, or workflow rules → load `sentry-create-alert`.
+
+## Dynamic Skill Fetching
+
+If the matched skill is **not installed locally**, fetch it directly from the repository:
+
+1. Build the raw URL: `https://raw.githubusercontent.com/getsentry/sentry-for-ai/main/skills/<skill-name>/SKILL.md`
+2. Fetch the file contents (e.g. `curl -sL <url>`) and follow its instructions.
+
+The [Skill Tree](../../SKILL_TREE.md) lists every available skill with its path — use it as the source of truth for valid skill names and paths.

skills/sentry-sdk-setup/SKILL.md

@@ -30,3 +30,12 @@ Detect the user's platform from project files (e.g. `package.json`, `go.mod`, `G
 3. If Next.js is detected, prefer `sentry-nextjs-sdk` over `sentry-react-sdk`.
 4. If React Native is detected, prefer `sentry-react-native-sdk` over `sentry-react-sdk`.
 5. If no match is found, direct the user to https://docs.sentry.io/platforms/ to find their platform.
+
+## Dynamic Skill Fetching
+
+If the matched SDK skill is **not installed locally**, fetch it directly from the repository:
+
+1. Build the raw URL: `https://raw.githubusercontent.com/getsentry/sentry-for-ai/main/skills/<skill-name>/SKILL.md`
+2. Fetch the file contents (e.g. `curl -sL <url>`) and follow its instructions.
+
+The [Skill Tree](../../SKILL_TREE.md) lists every available skill with its path — use it as the source of truth for valid skill names and paths.

skills/sentry-workflow/SKILL.md

@@ -23,3 +23,12 @@ Match the user's intent to the appropriate workflow skill below.
 2. If the user mentions Sentry bot comments, PR review comments from Sentry, or `sentry[bot]` → load `sentry-code-review`.
 3. If the user mentions Seer, bug prediction, or reviewing PRs for predicted issues → load `sentry-pr-code-review`.
 4. When unclear, ask whether the task involves live production issues (MCP) or PR review comments.
+
+## Dynamic Skill Fetching
+
+If the matched skill is **not installed locally**, fetch it directly from the repository:
+
+1. Build the raw URL: `https://raw.githubusercontent.com/getsentry/sentry-for-ai/main/skills/<skill-name>/SKILL.md`
+2. Fetch the file contents (e.g. `curl -sL <url>`) and follow its instructions.
+
+The [Skill Tree](../../SKILL_TREE.md) lists every available skill with its path — use it as the source of truth for valid skill names and paths.