issue-tracker

Instructions for managing issues and specifications stored as Markdown files in a repository’s .scratch directory, optionally through FrontierMCP.

In plain words
What is it for?
It supports reading and updating tickets, specifications, and their dependency graph, using the MCP server when available or files directly when it is not.
Why use it?
It gives agents one defined vocabulary and file format for understanding work items, relationships, and which tasks are available.

Agent

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 agents/51ck/frontier-mcp/issue-tracker
Clone the repo
git clone --depth 1 https://github.com/51ck/frontier-mcp
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,355 The whole file, excluding the scripts and references it only reads on demand.
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 $0.00000 $0.02355
Opus 5 $0.00000 $0.01177
Sonnet 5 $0.00000 $0.00471
Haiku 4.5 $0.00000 $0.00235

Measured yesterday against content hash c9c0d4b69b1b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

issue-tracker 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 yesterday.

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.

docs/agents/issue-tracker.md · 169 lines

How it starts

The opening of the file, as written. The whole thing — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Issue tracker — FrontierMCP and file fallback

Issues and specs live as markdown under .scratch/. Vocabulary is defined in the project's CONTEXT.md when present — Effort, Board, Header doc, Map, Spec, Ticket, Edge, Frontier.

When FrontierMCP (server name frontier) is loaded in your session, use the tools below. They own the schema, compute the graph, and answer orientation questions for a fraction of the token cost of reading every Ticket file. When it is not loaded, follow the File conventions literally — the same files, the same fields, written by hand.

Read this document once at setup. In Cursor it is also available as MCP resource frontier://tracker-doc.

Workspace

FrontierMCP resolves which repository it serves on every call:

  1. An explicit root argument on the call.
  2. The FRONTIER_ROOT environment variable.
  3. The server process working directory, walked upward to the nearest directory containing a .scratch/ directory, or .git in either form — a git worktree and a submodule carry .git as a file, and each is its own workspace rather than part of the repository enclosing it.

Register FrontierMCP once at user scope. Opening a project is the only per-repo setup step — no .cursor/mcp.json entry is required in each repository.

The working directory is the one your client launched the server in, and it is fixed for the session. Moving to another worktree mid-session does not retarget it: pass root on the call, set FRONTIER_ROOT, or restart the server. A worktree whose branch carries no .scratch/ of its own reports no Efforts — that is the worktree being served correctly, not the tracker going missing.

Every call that writes opens with root: <path>, naming the repository it just changed. Read it — a Ticket id is not repo-unique, so T10 updated looks the same whichever repository answered, and this line is the only part of the result that tells you. Reads do not carry it, and neither do the read paths of write-capable tools: edit_map with no section fields, spec with no content, and migrate_effort with preview.

Read the full file on GitHub · 169 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. yesterday First seen · 169 lines · 0 tokens per session scan A c9c0d4b69b1b

Subscribe to this mod's changes

issue-tracker is an agent published in the GitHub repository 51ck/frontier-mcp (0 stars, last pushed 13d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,355 tokens. 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.