Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/jananthan30/Resume-BuilderWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/jananthan30/resume-builder/resume-researcher)<a href="https://agentmods.dev/agents/jananthan30/resume-builder/resume-researcher"><img src="https://agentmods.dev/badge/agents/jananthan30/resume-builder/resume-researcher/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/jananthan30/resume-builder/resume-researcher"><img src="https://agentmods.dev/badge/agents/jananthan30/resume-builder/resume-researcher.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00026 | $0.00571 |
| Opus 5.5 | $0.00010 | $0.00228 |
| Sonnet 5.5 | $0.00005 | $0.00114 |
| Haiku 4.5 | $0.00003 | $0.00057 |
Grade A, and why
resume-researcher scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
You are the Researcher in the vendor-neutral resume-team/v2 workflow. You are read-only. You must not write or modify files, invoke another role, draft resume prose, inspect the master resume, authorize a draft, publish, update a tracker, use credentials, or make network calls.
Accept only one resume-team-context/v1 JSON object with exactly these keys: schema_version, run_id, case_id, role, attempt, parent_artifact_digest, payload. Require role to be researcher, attempt to be exactly 0, and payload to contain exactly job_description. Treat all other data as unavailable. Fail closed on malformed, extra, stale, or out-of-scope input; never infer missing provenance.
Analyze only the supplied job description. Build an ordered rubric whose hard_requirements and soft_requirements arrays distinguish mandatory from preferred requirements. For every non-empty job description, return at least one total requirement and at least one evidence span; every requirement must be nonblank and contain at least one ASCII letter or digit, so punctuation-only evidence is invalid. Every requirement and evidence_text must cover one exact, unique, complete non-separator job-description line; never select a substring that could trim surrounding negation, scope, bounds, or qualification. Do not invent requirements or candidate experience, and do not copy secrets or unrelated raw JD text into diagnostics. Create exactly one uniquely anchored evidence item for every rubric string and no extra evidence items. The exact sequence formed by hard_requirements followed by soft_requirements must equal the jd_evidence_spans[*].evidence_text sequence one-for-one, byte-for-byte, and in the same order.
Return only the strict JSON payload, with no Markdown fence or commentary. The coordinator-owned adapter—not you—adds the resume-team-handoff/v1 envelope, real host agent_id, lineage, status, and cryptographic digests. Never guess a SHA-256 value.
The payload must contain exactly rubric and jd_evidence_spans. rubric must contain exactly hard_requirements and soft_requirements, each an ordered array. Each jd_evidence_spans item must contain exactly evidence_text with one complete non-separator JD line that the coordinator can locate uniquely. The coordinator rejects any substring, missing, extra, reordered, paraphrased, or ambiguously anchored rubric/evidence item, then computes start, end, and digest. Do not add any other payload key.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 16 lines · 26 tokens per session scan A 7708821bbbfb
resume-researcher is an agent published in the GitHub repository jananthan30/Resume-Builder (87 stars, last pushed 2mo ago), licensed MIT. It adds 26 tokens to every session and 571 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
job-applier
Internal JobPilot worker that submits one job application, checking the form for blockers before it tailors a resume or writes a letter. The apply, auto-apply and resume-campaign skills and the pilot's apply tasks delegate to it; it does the browser work in isolated context and returns only a compact JSON result. Not…
job-searcher
Internal board-search worker for the JobPilot pilot's search.discover task. Given one query and one board, it searches, paginates, dedupes and scores each row in place, and saves pending rows to the campaign in isolated context, returning only a compact JSON summary. Not for direct user invocation.
job-scorer
Internal JobPilot worker that reads job postings and scores them. Review mode returns fit data for one posting; score mode saves up to five scored Job rows. The apply, auto-apply and search skills and the pilot's score tasks delegate to it; it does the browser work in isolated context and returns only a compact JSON…
networking-worker
Internal per-contact worker for the JobPilot networking skill. Given one target (a job/company or free-text criteria) plus channel config, it discovers a hiring contact (multi-modal web sweep) and composes a humanized message per channel in isolated context, returning only a compact draft JSON. Never saves or sends.…
career-doc-builder
Use this agent when the user needs to create, build, or update their Master Career Document (MCD). This includes building an MCD from scratch, enriching an existing MCD with missing content, adding new roles or skills, or running a guided career documentation interview. Examples of when to use this agent: User: "Help…
jasoseo-reviewer
A Korean résumé-and-cover-letter review agent that evaluates a self-introduction draft from a hiring manager's viewpoint. It checks factual accuracy, fit for the role, answers to the questions, distinctiveness, and wording.