Skip to content

Skill API Reference

Definitive reference for AgentOps SKILL.md frontmatter fields. Schema: schemas/skill-frontmatter.v1.schema.json.

Publicly, AgentOps talks about bookkeeping, validation, primitives, and flows. This document describes the internal API and taxonomy behind that operating model.

Frontmatter Format

Every skill has a YAML frontmatter block between --- delimiters at the top of SKILL.md:

YAML
---
name: my-skill
description: 'What this skill does. Triggers: "keyword1", "keyword2".'
skill_api_version: 1
context:
  window: fork
  intent:
    mode: task
  sections:
    exclude: [HISTORY]
  intel_scope: topic
metadata:
  tier: execution
---

Required Fields

Field Type Description
name string Skill identifier (must match directory name)
description string What the skill does, including trigger phrases
skill_api_version integer Always 1 (const)

Optional Fields

context

Controls what knowledge ao lookup --for=<skill> provides. Two forms:

String form (backward compat):

YAML
context: fork

Object form (recommended):

YAML
context:
  window: isolated
  intent:
    mode: task
  sections:
    exclude: [HISTORY]
  intel_scope: full

context.window

How the skill's execution context relates to the parent session.

Value Meaning
isolated Fresh context, no parent inheritance. For validation and mechanical skills.
fork Copy parent context as starting point. For skills that need to know what you're working on.
inherit Use full parent context as-is. For session utilities (status, handoff, recover).

v1 status: Parsed and stored. Not enforced at runtime (Phase 2).

context.sections

Filter which knowledge sections are injected.

YAML
sections:
  include: [INTEL, TASK]     # Allowlist — only these sections
  exclude: [HISTORY]         # Blocklist — everything except these

If both include and exclude are set, include takes precedence.

Valid section names:

Section Knowledge Fields
HISTORY Past session summaries
INTEL Learnings and patterns from the knowledge flywheel
TASK Current bead ID and predecessor context

v1 status: Actively enforced at runtime. ao lookup --for=<skill> zeroes excluded/non-included sections.

context.intent.mode

Declares what the skill is doing.

Value Meaning
task Executing work (implement, plan, validate)
questions Exploring or researching
none Operational utility (status, push, update)

v1 status: Parsed and stored. Not enforced at runtime (Phase 2 — orchestrators will use this to adapt behavior).

context.intel_scope

How much of the knowledge flywheel to inject.

Value Meaning
full All learnings and patterns
topic Only learnings matching the current query/task
none No learnings or patterns injected

v1 status: none is actively enforced (zeroes learnings + patterns). topic and full are declaration-only (Phase 2).

allowed-tools

Restricts which tools the skill can auto-approve.

YAML
# Array form
allowed-tools:
  - Read
  - Grep
  - Glob
  - Bash

# String form (comma-separated)
allowed-tools: Read, Grep, Glob, Bash

model

Preferred model for skill execution.

YAML
model: haiku    # Use cheaper/faster model for lightweight skills

Currently used by flywheel and status. Declaration-only — no CLI enforcement.

user-invocable

Whether the skill appears in the slash-command list.

YAML
user-invocable: true   # Shows as /skill-name
user-invocable: false  # Hidden from user, used by other skills

metadata

Skill classification and dependency information.

YAML
metadata:
  tier: execution           # See tier values below
  dependencies: [standards] # Skills loaded as context
  internal: false           # If true, not published externally
  version: "1.0.0"
  author: "Gas Town"
  triggers: ["keyword"]     # Additional trigger phrases
  replaces: old-skill-name  # Supersedes another skill

Tier Values

Tier Purpose Example Skills
judgment Legacy internal tier name for validation and review gates council, vibe, pre-mortem, post-mortem
execution Single-task implementation implement, bug-hunt, complexity, security-suite
orchestration Multi-skill coordination rpi, crank, swarm, evolve
session Session lifecycle handoff, recover, status, quickstart
background Mechanical utilities push, ratchet, flywheel, forge
knowledge Knowledge management compile, trace
product Product strategy product, readme, release, goals
library Shared references shared, standards, beads
meta System-level using-agentops, update, heal-skill
contribute External contributions pr-plan, pr-implement, pr-research, oss-docs
cross-vendor Cross-platform openai-docs, codex-team, converter, grafana-platform-dashboard

output_contract

Path to a JSON Schema file that defines the skill's structured output format.

YAML
output_contract: skills/council/schemas/verdict.json

Used by downstream skills to validate that upstream output matches the expected schema. Paths are relative to repo root.

v1 status: Declaration-only. No runtime enforcement. Consumed by skill chaining validation tooling.

Other Fields

Field Type Description
license string License identifier (e.g., MIT)
compatibility string Runtime requirements (e.g., Requires git, gh CLI)

Context Declaration Quick Reference

Core skills and their context policies:

Skill Window Sections Intent Intel Scope
Validation (tier: judgment)
council isolated exclude: HISTORY task full
vibe fork exclude: HISTORY task
pre-mortem fork exclude: HISTORY task
post-mortem fork exclude: HISTORY task
Orchestration
rpi fork
crank fork exclude: HISTORY task full
swarm fork exclude: HISTORY task full
evolve fork exclude: HISTORY task full
Execution
implement isolated exclude: HISTORY task topic
bug-hunt fork exclude: HISTORY task topic
doc fork exclude: HISTORY task topic
complexity fork exclude: HISTORY task topic
security fork exclude: HISTORY task topic
security-suite fork exclude: HISTORY task topic
reverse-engineer-rpi fork exclude: HISTORY task topic
grafana-platform-dashboard fork exclude: HISTORY, TASK questions none
Knowledge
research fork exclude: HISTORY, TASK questions topic
trace fork exclude: HISTORY task full
compile fork exclude: TASK task full
forge fork exclude: TASK task full
flywheel fork exclude: TASK task full
retro fork
Session
handoff inherit none none
recover inherit none none
status inherit none none
quickstart inherit none none
Background
push isolated exclude: HISTORY, INTEL, TASK none none
ratchet isolated exclude: HISTORY, INTEL, TASK none none
update isolated exclude: HISTORY, INTEL, TASK none none
heal-skill isolated exclude: HISTORY, INTEL, TASK none none
Product
product fork exclude: HISTORY task full
readme fork exclude: HISTORY task full
release fork exclude: HISTORY task full
goals fork exclude: HISTORY task topic
Contribute
pr-plan fork exclude: HISTORY task topic
pr-implement fork exclude: HISTORY task topic
pr-prep fork exclude: HISTORY task topic
pr-research fork exclude: HISTORY task topic
pr-retro fork exclude: HISTORY task topic
pr-validate fork exclude: HISTORY task topic
oss-docs fork exclude: HISTORY task topic
Library/Meta
shared isolated exclude: HISTORY, INTEL, TASK none none
standards isolated exclude: HISTORY, INTEL, TASK none none
using-agentops isolated exclude: HISTORY, INTEL, TASK none none
converter isolated exclude: HISTORY, INTEL, TASK none none
beads fork exclude: HISTORY task topic
inject fork
provenance fork
Cross-Vendor
openai-docs fork exclude: HISTORY, TASK questions none
codex-team fork
brainstorm inherit exclude: INTEL, HISTORY, TASK none none
plan fork task topic

Enforcement Summary (v1)

Field Runtime Enforcement
sections.include Active — zeroes non-included knowledge
sections.exclude Active — zeroes excluded knowledge
intel_scope: none Active — zeroes learnings + patterns
intel_scope: topic/full Declaration-only (Phase 2)
context.window Declaration-only (Phase 2)
context.intent.mode Declaration-only (Phase 2)
allowed-tools Active — controls auto-approval
model Declaration-only

See Also

  • Skills Reference — Skill descriptions and router
  • Skill Tiers — Taxonomy and dependency graph
  • Schema: schemas/skill-frontmatter.v1.schema.json