moonlighter
Job hunting from a Claude conversation: scans six ATS job boards, scores every posting against your profile with an LLM, and composes complete application answer sheets — you review and submit. Self-hosted MCP server; your data stays on your machine.
Emerging. No concerning findings. Grades remain capped until the project builds reputation through adoption. Grades reflect the full trust pyramid: code, verification depth, and reputation. New projects cap at C until adoption is earned.
How we verified
Verified is a snapshot. Live keeps it current, and builds your track record.
⚡ Connect GitHub → continuous verification on every pushwhy connect →Who stands behind it
albertosca
Source: github_repo_search
Claim it to get a verified publisher badge, a free copy of our full audit findings, and direct contact for any high-priority issues we find. Or connect your repo for our deepest verification, Live Monitored: read-only, revoke anytime. What we access →
Install from
The grade above is for the source repository. Registries can serve a different version, so we mark the ones we were not able to read.
These names and descriptions are the publisher's own, read from the source code. We print them as written. Our assessment is the findings above, not this list.
scan_and_evaluateScan job boards, evaluate with LLM, return new jobs above threshold.
scan_companyScan every open posting at ONE company right now and evaluate the new ones.
add_jobEvaluates a manually provided job and saves it to the database.
verify_jobRe-evaluates a job that couldn't be scored automatically because its description was empty, using text you copy from its real page.
archive_stale_jobsDetect and archive (status='archived', closed_at set) jobs that disappeared from their source.
list_jobsList jobs from DB filtered by status.
get_jobGet full details of a job posting.
prepare_applicationProduce the full set of answers for a job application, for you to paste into the form yourself. Reads the questions from the ATS API when it publishes them (Greenhouse, Recruitee); otherwise asks you to copy the page. job_id: ID of the job
prepare_application_from_pasteSame as prepare_application, for a page whose questions no API publishes: select the whole application page, copy it, and pass the text here. job_id: ID of the job page_text: everything copied off the application page
get_pipelineShow full application funnel: counts and list by status.
update_statusUpdate application status manually. status: 'screening' | 'interview' | 'offer' | 'rejected' | 'submitted' | 'draft' notes: free text notes appended to history next_action: e.g. 'follow up on 2026-06-01'
list_answer_bankEvery banked screening answer, most recently used first. Entries older than answer_bank_max_age_days are marked expired: kept, but no longer replayed on a sheet until a new submission refreshes them.
forget_answerDelete one banked answer by its question text (matched the way the bank keys it: lowercase, collapsed whitespace, no trailing punctuation). Use it when list_answer_bank shows an answer that is wrong or stale; the next application that asks the question goes to the LLM again.
setup_emailConfigure Gmail authentication for the account that receives application replies. Run only once. Opens the browser to authorize access. Requires the OAuth client file at email.credentials_path (default MOONLIGHTER_HOME/gmail-client.json) and writes the token to email.token_path — overwriting whateve…
sync_email_responsesRead recent emails in the configured Gmail account, whether read or unread, classify them with the LLM, and update the applications database. Returns a summary of the updates made.
MOONLIGHTER_HOMEThe wizard writes config.yaml into (defaults to ~/.moonlighter/). TwoANTHROPIC_API_KEYllm_backend: api — the Anthropic SDK, billed to API credits. RequiresMOONLIGHTER_DB_PATHTool annotations
No tools have read-only/destructive annotations
Add readOnlyHint or destructiveHint annotations to every tool so hosts can warn users before invoking.
All four hints declared on every tool
15/15 tools missing one or more hints — scan_and_evaluate (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); scan_company (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); add_job (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +12 more. OpenAI's directory rejects tools where any of the four hints are missing or non-boolean.
For every tool, set all four hints (readOnlyHint, destructiveHint, idempotentHint, openWorldHint) to explicit true/false values that match the handler’s actual behaviour.
Tool inputs are validated
11/15 tool handlers declare input schemas (73%)
Declare an inputSchema with zod/joi/yup on every tool definition.
Claim the listing to review these findings one by one and send us a correction where you disagree, straight to the team. Claiming also means we tell you when the grade moves, and reach you first if we find anything urgent.
[](https://m8ven.ai/mcp/albertosca/moonlighter)?variant=verified to the badge URL.Vetting this one by hand? Tool Check is an MCP that scores other MCPs. Add it once and ask Claude, ChatGPT, or any MCP client to grade a server, surface CVEs, check the publisher, and suggest safer alternatives — before you install.
https://m8ven.ai/api/mcp/tool-check