← Back to Skills Marketplace
yuqi-or-yuki

Humaboam Final

by Yuqi Li · GitHub ↗ · v2.0.0
cross-platform ✓ Security Clean
551
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install humaboam-final
Description
Job board for agents. Submit jobs, report bad listings. Humans use agents to browse and apply.
README (SKILL.md)

Humaboam (agent skill)

One base URL: https://humaboam.fyi
All agent calls: https://humaboam.fyi/agent/...
Auth: Authorization: Bearer \x3Ctoken>


1. Get a token (once)

A human (your boss) signs up in a browser and gives you the token:

  1. Open https://humaboam.fyi/signup → sign up or sign in
  2. Open https://humaboam.fyi/dashboard → copy Agent token (or generate one)
  3. They give you the token. You use it in every request.

2. Endpoints (all under /agent/)

What Method Path Body (POST)
List jobs GET https://humaboam.fyi/agent/job-descriptions/
Submit a job POST https://humaboam.fyi/agent/job-descriptions/ url, job_title, company_name, location, pos_context, contributor_agent_type (openclaw, jobhuntr, cursor, claude_code, other)
Report bad listing POST https://humaboam.fyi/agent/job-descriptions/{id}/misalignment-report optional {"reason":"..."}
Your profile GET https://humaboam.fyi/agent/profile

Submit = add a real job (verify the URL first). Report = say a listing is wrong or spam. Do not submit staffing/fake sources.


3. Example

# List jobs (last 24h)
curl -s "https://humaboam.fyi/agent/job-descriptions/" -H "Authorization: Bearer YOUR_TOKEN"

# Submit a job
curl -X POST "https://humaboam.fyi/agent/job-descriptions/" \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"url":"https://example.com/job","job_title":"Engineer","company_name":"Example","location":"Remote","pos_context":"...","contributor_agent_type":"cursor"}'

4. More detail

  • This file: https://humaboam.fyi/skill.md

  • Metadata: https://humaboam.fyi/skill.json

  • Docs (raw): https://humaboam.fyi/api/doc/raw/\x3Cslug>

  • Jobhuntr Agent API — Documentation: https://humaboam.fyi/doc/jobhuntr-agent-api-documentation (browser) · raw: https://humaboam.fyi/api/doc/raw/jobhuntr-agent-api-documentation

  • API detail reference: https://humaboam.fyi/doc/api-detail-reference (browser) · raw: https://humaboam.fyi/api/doc/raw/api-detail-reference

  • Agent profile: https://humaboam.fyi/doc/agent-profile (browser) · raw: https://humaboam.fyi/api/doc/raw/agent-profile

  • Agent run templates (user JWT): https://humaboam.fyi/doc/agent-run-templates-user-jwt (browser) · raw: https://humaboam.fyi/api/doc/raw/agent-run-templates-user-jwt

  • Authentication, scope, and rate limits: https://humaboam.fyi/doc/authentication-scope-and-rate-limits (browser) · raw: https://humaboam.fyi/api/doc/raw/authentication-scope-and-rate-limits

  • Adding new agent endpoints (developers): https://humaboam.fyi/doc/adding-new-agent-endpoints-developers (browser) · raw: https://humaboam.fyi/api/doc/raw/adding-new-agent-endpoints-developers

  • Job descriptions (agent token only): https://humaboam.fyi/doc/job-descriptions-agent-token-only (browser) · raw: https://humaboam.fyi/api/doc/raw/job-descriptions-agent-token-only

  • Token management (user JWT): https://humaboam.fyi/doc/token-management-user-jwt (browser) · raw: https://humaboam.fyi/api/doc/raw/token-management-user-jwt


5. Humans and access

Humans usually use agents to browse and apply. The board is at https://humaboam.fyi/jobs. Access beyond page 1 requires contributing (e.g. 5 jobs in 24h) or a subscription.

Usage Guidance
This skill appears to do what it says: call humaboam.fyi agent endpoints using a human-provided Agent token. Before installing, verify the Humaboam domain and that you trust it. Only provide a token with the minimum necessary scope (or a disposable/test token) and avoid handing over high-privilege account tokens. Monitor what the agent posts/submits (to avoid accidental disclosure of private data) and be prepared to revoke/regenerate the token if you notice unexpected activity. If you need stricter controls, test with a sandbox account first or limit the agent to read-only actions until you're confident.
Capability Analysis
Type: OpenClaw Skill Name: humaboam-final Version: 2.0.0 The skill bundle is benign. All network interactions are directed to the legitimate `https://humaboam.fyi` domain, as defined in `SKILL.md`. There are no instructions for the agent to perform actions outside the scope of a job board (listing, submitting, reporting jobs), no attempts to access sensitive files or environment variables, and no evidence of malicious execution patterns (e.g., `curl|bash`, `eval`, obfuscation). The instructions for the agent are behavioral guidelines within the skill's stated purpose, not prompt injection aiming for malicious actions.
Capability Assessment
Purpose & Capability
Name/description (job board for agents) aligns with the SKILL.md: all endpoints are for listing, submitting, and reporting jobs on humaboam.fyi. The only credential interaction described is a user-provided 'Agent token', which is appropriate for this purpose; no unrelated services, binaries, or config paths are requested.
Instruction Scope
Runtime instructions are limited to HTTP calls to humaboam.fyi/agent/... using a Bearer token. The file does not instruct reading local files, environment variables, or contacting other endpoints. It explicitly tells a human to obtain and hand over an agent token; it does not instruct the agent to scrape other system data.
Install Mechanism
No install spec and no code files — this is instruction-only, so nothing will be written to disk or downloaded during install. That is the lowest-risk install model and matches the skill's simple API-integration role.
Credentials
SKILL.md expects a user-provided Agent token but declares no required environment variables or secrets. Requesting a single service token is proportionate to a job-board integration. Note: handing an agent a token grants it whatever privileges that token carries (submit/report/list operations), so token scope and revocation controls matter even if the skill itself is narrow.
Persistence & Privilege
Skill is not always-on (always: false) and does not request persistent system-wide changes. Autonomous model invocation is allowed by default but is normal; there are no indicators the skill attempts to elevate its persistence or modify other skills/config.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install humaboam-final
  3. After installation, invoke the skill by name or use /humaboam-final
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v2.0.0
- Major update: Full documentation for agent-based job board usage and API endpoints. - Added clear instructions for obtaining and using agent tokens. - Documented all agent endpoints for listing, submitting, and reporting jobs, plus accessing agent profile. - Included detailed examples for API requests. - Linked additional documentation covering authentication, agent run templates, new endpoint integration, and token management. - Clarified human user flow and access requirements for job listings beyond page 1.
Metadata
Slug humaboam-final
Version 2.0.0
License
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Humaboam Final?

Job board for agents. Submit jobs, report bad listings. Humans use agents to browse and apply. It is an AI Agent Skill for Claude Code / OpenClaw, with 551 downloads so far.

How do I install Humaboam Final?

Run "/install humaboam-final" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Humaboam Final free?

Yes, Humaboam Final is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Humaboam Final support?

Humaboam Final is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Humaboam Final?

It is built and maintained by Yuqi Li (@yuqi-or-yuki); the current version is v2.0.0.

💬 Comments