AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Goat

skill-blundergoat-goat-flow-goat · by blundergoat

Use when you describe an outcome and need the right goat-* workflow chosen for you.

No reviews yet
0 installs
35 views
0.0% view→install

Install

$ agentstack add skill-blundergoat-goat-flow-goat

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-blundergoat-goat-flow-goat)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
3mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Goat? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

/goat

Shared Conventions

Read .goat-flow/skill-docs/skill-preamble.md for shared conventions.

Use when the user gives an outcome and needs the right goat-* route. If the user names a skill explicitly (/goat-debug, /goat-review, etc.), route immediately - no classification, no GATHER.

If a symptom tempts code reading, STOP. The dispatcher routes; the routed skill investigates.

| Excuse | Reality | |--------|---------| | "I can see it - routing is overhead" | You are dispatcher, not investigator. Route first. | | "The user said 'just fix it'" | Pressure is not an override. Route to /goat-debug. | | "Time pressure means investigate now" | Routing takes seconds; wrong routing wastes more. | | "Multiple symptoms mean read files" | Split numbered intents; route each separately. |

How It Works

  1. UNDERSTAND - classify intent and target. If multiple intents, number each and route independently. Ask only if ordering matters.
  2. GATHER - before routing, check:
  • Footgun matches: grep .goat-flow/learning-loop/footguns/INDEX.md for the target area; open entries only on hits
  • Ask-first boundaries: scan the active instruction file's Ask First boundaries for named files; if none are named, record target-files=unknown
  • If any check fails or is unavailable, note gather-degraded and route anyway
  • Do not emit the preamble's Relevant prior learnings line - that belongs to the routed skill's Step 0
  1. ROUTE - dispatch using the route map. Emit a Route Snapshot (Intent / Route / Rationale), e.g.:
Intent: Diagnose a slow endpoint
Route: /goat-debug
Rationale: "slow" is a symptom to investigate; no file named -> target-files=unknown

Route Map

| Intent | Route | |--------|-------| | Bug, failure, unexpected behaviour | /goat-debug | | Verify a fix worked | /goat-debug (post-fix verification) | | Browser-visible issue | Browser evidence first; /goat-debug Investigate if diagnosis needed | | Understand, explain, explore unfamiliar code | /goat-debug (Investigate mode) | | Quality review, audit, diff check | /goat-review | | Verify a diff/PR before merge | /goat-review | | Multi-perspective critique | /goat-critique | | Security, compliance, dependency audit | /goat-security | | Testing gaps, coverage, verification planning | /goat-qa | | Verify test coverage | /goat-qa | | Feature planning, milestones | /goat-plan | | Bare task path (no action verb) | Bare or ambiguous task paths are read-only context. Do not update .active, milestone status, or code from a path alone | | Build/plan verb + scope | /goat-plan (Step 0 handles complexity and mode) | | Simple implementation (single-file, obvious) | No skill; use execution loop directly | | Simple question | Answer directly |

More examples: /goat-review this diff -> /goat-review (explicit; no GATHER). Look at auth -> /goat-security (assume security audit; offer /goat-review re-route). Debug login test then review fix -> 1. /goat-debug; 2. /goat-review.

Constraints

  • MUST respect explicit skill invocations immediately - no reclassification
  • MUST NOT inspect source code, read implementation files, or make changes before routing
  • MUST understand intent conversationally, not via keyword lookup - 0-2 clarification questions max; route with stated assumption if still ambiguous
  • MUST emit a Route Snapshot with every dispatch - Proof Gate applies to route claims
  • MUST split multi-intent requests into numbered intents and route each
  • MUST pass brief/depth to target skill and preserve context on re-route

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.