agent-forge
Design an agent only when a durable autonomy boundary is justified; compile it only when a Foundry runtime is available.
Skill by frankxai
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/frankxai-agent-forge
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: design, backend, agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/agent-forge
What's inside
- Skill name
agent-forge- Description
- Design an agent only when a durable autonomy boundary is justified; compile it only when a Foundry runtime is available. Use for persistent decision rights, distinct memory, constrained tools, ownership transfer, or an ongoing schedule, channel, or API trigger.
- License
- MIT
- Instructions
- 68 lines in SKILL.md · view on GitHub
Files (3)
- SKILL.md
- openai.yaml · agents/
- portable-contracts.md · references/
SKILL.md (first lines)
# Agent Forge
## Outcome
Design the smallest justified persistent actor, or deliberately downgrade the request to a skill or temporary worker. Compile it only when a Foundry runtime is present.
## Agent necessity gate
An agent must own at least one durable boundary:
- stable decision rights;
- a distinct memory scope;
- a constrained tool or permission boundary;
- genuine ownership transfer or handoff authority;
- an ongoing schedule, channel, or API trigger.
If none applies, stop and route to `$skill-forge`. A name, persona, or desire for “more intelligence” is not an agent boundary.
## Procedure
1. Build the Task Envelope.
- Record the objective, deliverables, stakes, autonomy, permissions, evidence policy, and termination tests.
2. Prove necessity.
- Mark each necessity-gate field true or false.
- Explain the persistent boundary in operational language.
- Compare the proposed agent against a skill, deterministic tool, and temporary task worker.
3. Define decision rights.
- State what the agent may decide, what it may recommend, and what requires operator approval.
- Exclude public release, financial commitment, destructive action, permission expansion, and substrate change unless explicitly authorized.
4. Define memory and tools.
- Name readable and writable scopes separately.
- Choose turn, session, or durable retention.
- Allow only the minimum tools needed; list denied tools explicitly.
5. Define handoffs and termination.
- Give every handoff a condition and target.
- Set success conditions, stop conditions, and a hard turn or round ceiling.
6. Compose existing skills.
- Agents orchestrate capabilities; they do not duplicate skill procedures inside a persona prompt.
7. Select the execution mode.
- **Foundry runtime available:** only when `tools/foundry/cli.mjs` and `foundry/contracts/` exist in the current workspace, compile and prove:
```bash
node tools/foundry/cli.mjs forge \
--envelope <task-envelope.json> \
--pack <agent-pack.json> \
--out <package-directory>
node tools/foundry/cli.mjs prove <package-directory>
```
- **Portable skills-only install:** read `references/portable-contracts.md`, then author the Task Envelope and Agent Pack with those field names. Mark compilation and proof `pending-runtime`; do not invent an Evidence Receipt. Provide the commands above as an optional handoff for a later SIS workspace run.
8. Refuse false persistence.
- If the boundary disappears during design, return a Skill Pack proposal instead of forcing agent creation.
## Quality gates
- At least one durable boundary is true and explained.
- Decision rights do not exceed the Task Envelope autonomy.
- Tool and memory permissions are least-privilege.
- Handoffs do not form an unbounded loop.
- The agent has explicit success and stop conditions.
- Behavioral, security, and drift evidence are required before production autonomy.
## Return
Return the necessity verdict, chosen topology, Agent Pack, execution mode, permissions requiring approval, and any downgrade recommendation. Include a compiled package path and Evidence Receipt only when they were actually produced.From frankxai/Starlight-Intelligence-System, MIT.
More from frankxai/Starlight-Intelligence-System
- 0.588score
system-forge
frankxai
Design a multi-capability system as a swarm, vertical, or plugin; compile it only when a Foundry runtime is available.
SkillAFreshness: A★ 11Claude Code +9 - 0.588score
skill-forge
frankxai
Forge a reusable, portable skill from a proven workflow or capability brief.
SkillAFreshness: A★ 11Claude Code +9 - 0.588score
agent-hygiene
frankxai
The quality bar for any agent working on agent definitions, skills, commands, hooks, AGENTS.md or CLAUDE.md, in any repo and any harness.
SkillAFreshness: A★ 11Claude Code +9 - 0.585score
taste-engine
frankxai
Turn subjective quality intent into executable hard gates, weighted rubrics, exemplars, anti-exemplars, blind comparison, and independent judge evidence.
SkillAFreshness: A★ 11Claude Code +9 - 0.584score
starlight-agent-designer
frankxai
Design or improve Starlight agents from a value opportunity through architecture, agent contracts, prompt engineering, bounded proactivity and evaluation.
SkillAFreshness: A★ 11Claude Code +9 - 0.583score
estate-guard
frankxai
Security gate for agentic repos.
SkillAFreshness: A★ 11Claude Code +9 - 0.578score
repo-mastery
frankxai
Master ACOS, FrankX, Starlight-Intelligence-System, arcanea, claude-code-config and full starlight/repos ecosystem.
SkillAFreshness: A★ 11Claude Code +9 - 0.578score
harness-integration
frankxai
ACOS Grok bridge.
SkillAFreshness: A★ 11Claude Code +9 - 0.572score
creative-director-draft
frankxai
Compile, inspect and revise the Starlight creative-director draft example without provider execution.
SkillAFreshness: A★ 11Claude Code +9 - 0.551score
community-loop
frankxai
Plan or review a small creation cell's weekly quest, member commitments, artifacts, peer help, and reflections.
SkillBFreshness: B★ 11Claude Code +9 - 0.551score
starlight-web-production
frankxai
Direct, build, redesign, or audit high-value websites, landing pages, product UI, design systems, and web releases for FrankX, Starlight, Arcanea, GenCreator…
SkillBFreshness: B★ 11Claude Code +9 - 0.541score
starlight-command-center
frankxai
Inspect a venture portfolio, identify operating pressure, and render a Starlight command center when the user asks for status, priorities, risks, or a visual…
SkillBFreshness: B★ 11Claude Code +9
Signals
- Stars
- 11
- Forks
- 3
- Open issues
- 61
- Last commit (this folder)
- Oct 10, 2026 (18 days ago)
- Last release
- Jun 10, 2026 (4 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
Same skill elsewhere
Copies of this SKILL.md published by others; search shows only one of them.
- 0.340score
agent-forge
Arcanea-Labs
Design an agent only when a durable autonomy boundary is justified; compile it only when a Foundry runtime is available.
SkillAFreshness: AClaude Code +9 - 0.329score
agent-forge
frankxai
Design an agent only when a durable autonomy boundary is justified; compile it only when a Foundry runtime is available.
SkillBFreshness: B★ 11Claude Code +9 - 0.329score
agent-forge
frankxai
Design an agent only when a durable autonomy boundary is justified; compile it only when a Foundry runtime is available.
SkillBFreshness: B★ 11Claude Code +9 - 0.316score
agent-forge
Arcanea-Labs
Design an agent only when a durable autonomy boundary is justified; compile it only when a Foundry runtime is available.
SkillBFreshness: BClaude Code +9
More from frankxai
- 0.618score
remotion-video-agent
frankxai
Use when creating or editing Remotion videos, programmatic explainers, motion graphics, MP4 renders, GIF renders, social cutdowns, or React-authored video…
SkillAFreshness: A★ 78Claude Code +9 - 0.618score
web-motion-auditor
frankxai
Use when reviewing web animations, UI transitions, scroll effects, hover states, loading states, GIF/video captures, reduced-motion support, animation…
SkillAFreshness: A★ 78Claude Code +9 - 0.618score
generative-media-director
frankxai
Use when planning, prompting, generating, editing, routing, finishing, or QAing AI-created images, GIFs, videos, style frames, image-to-video clips, social…
SkillAFreshness: A★ 78Claude Code +9 - 0.618score
motion-director
frankxai
Use when designing motion concepts, animation systems, choreography, timing, easing, reduced-motion behavior, or deciding which motion stack to use for web…
SkillAFreshness: A★ 78Claude Code +9 - 0.614score
site-motion-excellence-director
frankxai
Use when planning, auditing, specifying, implementing, or improving premium website/app motion across FrankX, Arcanea, Starlight, or Vercel/Next.js sites.
SkillAFreshness: A★ 78Claude Code +9 - 0.588score
agent-hygiene
frankxai
The quality bar for any agent working on agent definitions, skills, commands, hooks, AGENTS.md or CLAUDE.md, in any repo and any harness.
SkillAFreshness: A★ 11Claude Code +9
Cite this page
agent-forge skill. Agent Mart. https://agentmart.dev/skills/frankxai-agent-forge (last updated 2026-10-24)