CtrlK
BlogDocsLog inGet started
Tessl Logo

farcaster-agent

Create Farcaster accounts and post casts autonomously. Official skill from the Farcaster team.

Install with Tessl CLI

npx tessl i github:teeclaw/phorm-nft --skill farcaster-agent
What are skills?

71

1.18x

Quality

58%

Does it follow best practices?

Impact

96%

1.18x

Average score across 3 eval scenarios

Optimize this skill with Tessl

npx tessl skill review --optimize ./skills/farcaster-agent/SKILL.md
SKILL.md
Review
Evals

Evaluation results

89%

39%

Farcaster Account Bootstrap for an Autonomous AI Agent

Account setup workflow

Criteria
Without context
With context

Ethers wallet creation

0%

100%

Prints wallet address

100%

100%

Prints private key

100%

100%

Auto-setup invocation

0%

33%

PRIVATE_KEY env var

50%

70%

Funding amount guidance

37%

100%

Multi-chain funding mention

0%

100%

Credential storage documented

100%

100%

No --no-save flag

100%

100%

DiceBear or HTTPS PFP hint

0%

100%

README produced

100%

100%

Without context: $1.0580 · 4m 16s · 58 turns · 135 in / 13,213 out tokens

With context: $2.2714 · 6m 58s · 76 turns · 329 in / 22,032 out tokens

100%

Scheduled Cast Publisher for Farcaster

Cast posting with credentials

Criteria
Without context
With context

loadCredentials import

100%

100%

postCast import

100%

100%

custodyPrivateKey field

100%

100%

signerPrivateKey field

100%

100%

fid as Number

100%

100%

Correct cast URL format

100%

100%

Logs cast URL per cast

100%

100%

Sequential posting

100%

100%

No hardcoded keys

100%

100%

plan.md credential fields

100%

100%

Without context: $1.6190 · 5m 3s · 60 turns · 909 in / 12,415 out tokens

With context: $1.0065 · 3m 54s · 45 turns · 876 in / 11,350 out tokens

100%

5%

Farcaster Profile Configurator for AI Agents

Profile setup and fname validation

Criteria
Without context
With context

Lowercase only check

100%

100%

No leading hyphen check

100%

100%

Max 16 chars check

100%

100%

Trailing hyphen / invalid char

100%

100%

Valid names accepted

100%

100%

DiceBear PFP URL

100%

100%

setupFullProfile params

58%

100%

Env var credential loading

100%

100%

setupFullProfile import

100%

100%

profile-config-report.json

100%

100%

Validation reason in report

100%

100%

Without context: $0.9202 · 3m 37s · 47 turns · 184 in / 12,005 out tokens

With context: $1.1063 · 4m 42s · 46 turns · 79 in / 14,904 out tokens

Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.