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.
npx skills add strelov1/freehire-cli --skill freehire-mail-triagegit clone --depth 1 https://github.com/strelov1/freehire-cliWrote 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/skills/strelov1/freehire-cli/freehire-mail-triage)<a href="https://agentmods.dev/skills/strelov1/freehire-cli/freehire-mail-triage"><img src="https://agentmods.dev/badge/skills/strelov1/freehire-cli/freehire-mail-triage/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/skills/strelov1/freehire-cli/freehire-mail-triage"><img src="https://agentmods.dev/badge/skills/strelov1/freehire-cli/freehire-mail-triage.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 | $0.00096 | $0.01513 |
| Opus 5.5 | $0.00038 | $0.00605 |
| Sonnet 5.5 | $0.00019 | $0.00303 |
| Haiku 4.5 | $0.00010 | $0.00151 |
Grade A, and why
freehire-mail-triage 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.
How it starts
The opening of the file, as written. The whole thing — 119 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sorting application mail into the tracker
Needs a key: freehire auth login --token fhk_… (or FREEHIRE_TOKEN).
freehire does not fetch mail. You bring your own client — himalaya, mbsync,
notmuch, the Gmail API, anything that can emit JSON — and hand the result over.
In exchange you get the inbox wired into the application tracker without paying for
a mail connector or a classifier.
The loop is four steps, and then two queues that only a human's verdict can empty:
# 1. Your client fetches; you shape it into the batch and push it.
himalaya envelope list --output json | jq '[.[] | {
external_id: .message_id, from_addr: .from.addr, from_name: .from.name,
subject: .subject, received_at: .date }]' | freehire inbox push
# 2. Ask for what still needs judging, with the text to judge.
freehire --json inbox list --unclassified --body --limit 50
# 3. You decide what each message is (and which application it belongs to).
# 4. Record the verdict — one call sets the label, the link, and the stage.
freehire inbox triage 812 rejection --slug go-dev-acme-t35nijto
freehire inbox triage 813 interview_invitation --slug data-eng-globex-9f2k1a0z --confidence 0.9
external_id is the deduplication key — use the message's Message-ID. Pushing
the same id again updates that message rather than storing a copy, so re-running
your sync over an overlapping window is safe and cheap. It never un-reads a message,
never resurrects one the user deleted, and never overwrites a verdict you recorded.
Batches are capped at 100 messages.
Use inbox list --body, not inbox read, when sweeping. Opening a message marks
it read, and read is meant for "the user asked to see this one". Sweeping a backlog
with read would silently zero the user's unread count. The listing returns the same
readable text (HTML-only ATS mail arrives stripped to text) and marks nothing.
Signals — acknowledgement, screening, interview_invitation, assessment,
offer, rejection, info_request, incomplete_application, other. A forward
signal on a linked message advances that application's stage; a settled application
(rejected/accepted/withdrawn) is never dragged back into the pipeline.
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 · 119 lines · 96 tokens per session scan A 0cc38f89f177
freehire-mail-triage is a skill published in the GitHub repository strelov1/freehire-cli (28 stars, last pushed 2d ago), licensed MIT. It adds 96 tokens to every session and 1,513 once invoked, about $0.0004 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 2026-08-31.
Other skills, from other repositories
resume-tailoring
Methodology for tailoring a resume to a job description honestly and effectively — fit analysis, gap handling, keyword mirroring, fast-learner framing, and primary-over-secondary emphasis. Use whenever curating or rewriting a resume for a specific role.
mentor-patterns
The six patterns the Mentor hunts in the records to decide WHEN to speak. Silence is the default; only a real, recurring pattern earns a word. This skill gives the canonical detection method for each pattern (DB query + threshold) so the Mentor never speaks from a single data point. Read-only — never writes to the DB.…
onboarding-flow
Conversational protocol the Assistente follows to onboard the user — first message, iterative one-question-per-turn pacing, blocking checklist (the floor that unlocks the dashboard) vs rich checklist (what makes Writers actually useful), settore-agnostic question style (NEVER assume IT), and the mandatory checkpoint…
apply-flow
How the CLOSER runs one authorised application with applyflow.py — the checkpointed state machine (detect, fill, uploadcv, screening, review, submit), the mandatory receipt without which applied is never written, and what to do on each result, blockedhuman first of all. Use it for every position taken from the queue.…
alive:system-cleanup
The world feels messy. Stale tasks, orphan folders, v2 remnants, unsaved sessions — entropy is accumulating and needs to be addressed before it compounds. Scans across all walnuts, then surfaces issues one at a time.
application-flow
DB + filesystem contract every Scrittore follows when taking a position from scored (≥50) to ready/excluded. Three gates BEFORE writing a single line of CV (anti-rewriting, anti-collision, link verification), one canonical path for deliverables, one final gate after the 3rd Critic round. Skipping any of these produces…