trackly-cli CLAUDE.md

Repository instructions for Trackly CLI, a terminal and MCP job-search tool that searches jobs across companies for people using AI coding assistants.

In plain words
What is it for?
They guide the CLI and MCP server, Google sign-in setup, repository navigation, supported runtime requirements, and production data checks.
Why use it?
They tell agents which backend contains live production data and prevent them from using outdated databases or unsafe direct access methods.

Instructions file

Install

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.

agentmods
npx agentmods add instructions/trackly-app/trackly-cli/claude-md
Clone the repo
git clone --depth 1 https://github.com/trackly-app/trackly-cli
Per session 3,195 This file is loaded in full into every session.
When invoked 3,195 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.03195 $0.03195
Opus 5 $0.01597 $0.01597
Sonnet 5 $0.00639 $0.00639
Haiku 4.5 $0.00319 $0.00319

Measured 2d ago against content hash b45594f93d81, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

trackly-cli CLAUDE.md 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 2d 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.

CLAUDE.md · 159 lines

How it starts

The opening of the file, as written. The whole thing — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.

trackly-cli

CLI + MCP server for the Trackly job tracker. Lets users search 170K+ jobs across 3,800+ companies from the terminal or through AI agents (Claude Code, Cursor) via MCP.

Tech Stack

  • Runtime: Node.js ^20.20.0 || >=22.22.0 (pure JS, no build step, no TypeScript)
  • MCP SDK: @modelcontextprotocol/sdk (stdio transport)
  • Validation: zod (MCP tool input schemas)
  • Auth: Google OAuth via local callback server, tokens stored in ~/.trackly/config.json
  • API: All requests go to https://closeai.mba (the backend, same as CLOSE AI)

Backend Production Source Of Truth

After the 2026-06-30 Azure cutover, https://closeai.mba is served by Azure and the live DB is Azure blue Postgres behind the backend/VNet. This CLI repo is a consumer only; do not use AWS RDS, Render, old DB aliases, ssh closeai-web, or direct SQL for live production claims, migrations, user exports, or company-add decisions. Backend data checks belong in protected close-ai admin/report endpoints, not this CLI repo.

Directory Structure

bin/trackly          # CLI entrypoint (shebang script). All CLI commands + arg parser + main()
lib/agent.js         # Agent setup, doctor, private resume cache, and public skill installation
lib/client.js        # HTTP client: config loading, token refresh, apiRequest()
lib/formatters.js    # Terminal output: color(), outputJobs(), outputCompanies(), outputStats(), outputContacts(), outputReferralCampaign(), outputNetworkBrief()
mcp/server.js        # MCP server entrypoint and search/network tools, launched via `trackly mcp`
mcp/apply-tools.js   # Trackly Apply MCP schemas, validators, and tool registration
contracts/           # Versioned hosted/local Trackly Apply MCP schema contract
skills/trackly-apply/  # Sanitized public browser-mechanics skill bundled with the CLI
scripts/             # Maintainer checks; the packaged audit verifier is the named exception
docs/trackly-tools.md  # MCP tool reference (for embedding in AI contexts)
docs/solutions/       # documented solutions organized by category with searchable YAML frontmatter (module, tags, problem_type)
CONCEPTS.md           # shared domain vocabulary for entities, named processes, and status concepts
server.json          # MCP Registry manifest (io.github.trackly-app/trackly)

Read the full file on GitHub · 159 lines

Changes

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.

  1. 2d ago First seen · 159 lines · 3,195 tokens per session scan A b45594f93d81

Subscribe to this mod's changes

trackly-cli CLAUDE.md is an instructions file published in the GitHub repository trackly-app/trackly-cli (3 stars, last pushed 3d ago), licensed MIT. It adds 3,195 tokens to every session, about $0.0160 per session on Opus 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.