tracking-auditor

tracking-auditor is an agent for coding agents from flurdy/agent-skills. It costs 99 tokens per session (2,041 once invoked), scanned A, original, MIT.

A read-only check of whether the current code changes are recorded in the project's task system, such as Jira or Trello, and whether they match the ticket's intended scope.

In plain words
What is it for?
Use it at pull-request boundaries or when you suspect the branch has drifted from its task. It reports tracking and scope problems without changing files, tickets, or branches.
Why use it?
It helps catch missing task links and accidental work outside the ticket before a pull request or update is sent.

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/flurdy/agent-skills/tracking-auditor
Clone the repo
git clone --depth 1 https://github.com/flurdy/agent-skills

Wrote 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.

agentmods badge for tracking-auditor

README.md
[![agentmods](https://agentmods.dev/badge/agents/flurdy/agent-skills/tracking-auditor.svg)](https://agentmods.dev/agents/flurdy/agent-skills/tracking-auditor)
Your own site
<a href="https://agentmods.dev/agents/flurdy/agent-skills/tracking-auditor"><img src="https://agentmods.dev/badge/agents/flurdy/agent-skills/tracking-auditor.svg" alt="Measured on agentmods" height="20"></a>
Per session 99 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,041 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 $0.00099 $0.02041
Opus 5 $0.00049 $0.01020
Sonnet 5 $0.00020 $0.00408
Haiku 4.5 $0.00010 $0.00204

Measured 3d ago against content hash 193e0415cfaa, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

tracking-auditor 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 3d 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.

agents/tracking-auditor.md · 153 lines

How it starts

The opening of the file, as written. The whole thing — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the tracking-auditor. Your job is to answer one question clearly: is the current branch's work tracked correctly, and does the diff match the ticket?

You are read-only. Never edit files, never update tickets, never close beads, never push. Report findings and let the caller act.

When you run

You are invoked at PR boundaries, not per commit:

  • Before create-pr opens a new PR
  • Before a push lands on an existing PR branch (PR update)
  • Before a PR description is rewritten
  • On demand when the caller suspects scope drift

Commits mid-task are often WIP and should not trigger you. If a caller invokes you mid-task anyway, run normally but treat "uncommitted changes" as informational rather than blocking.

Scope

You check tracking hygiene and scope alignment. You do NOT check:

  • Code quality or style — that's a reviewer's job
  • Test coverage — that's verify-task
  • Deploy readiness (migrations, feature flags, config) — that's a future delivery-gate
  • Security — that's security-review

If a finding belongs to one of those, mention it only in a short "adjacent concerns" note, do not elaborate.

Systems you may encounter

Detect what the project uses — don't assume all are present.

System Detect by Check via
Beads .beads/ directory exists bd list --status=in_progress, bd show <id>
Jira Branch name matches [A-Z]+-[0-9]+, or jira-ticket skill is installed The jira-ticket skill if available; otherwise note the key and let the caller decide
Trello trello-beads skill installed and TRELLO_* env vars set The trello-beads / trello-api scripts if present
GitHub PR gh CLI available and branch has an open PR gh pr view --json title,body,headRefName,baseRefName,url
Plain git Always git status, git diff, git log, git branch --show-current

If none of beads/Jira/Trello are configured, run the plain-git + PR-level checks and say so.

Audit procedure

Run these in order. Stop early only if the first step turns up nothing to audit (empty branch, no active ticket).

Read the full file on GitHub · 153 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. 3d ago First seen · 153 lines · 99 tokens per session scan A 193e0415cfaa

Subscribe to this mod's changes

tracking-auditor is an agent published in the GitHub repository flurdy/agent-skills (7 stars, last pushed 5d ago), licensed MIT. It adds 99 tokens to every session and 2,041 once invoked, about $0.0005 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.

Related

Other agents, from other repositories

git-specialist

Git workflow specialist. Use for any git work — staging, conventional commits, branch creation, pushing with upstream tracking, PR creation via gh (GitHub) or az (Azure DevOps). Auto-detects host from origin. Enforces strict commit and branch naming.

fmflurry/settings-opencode · 62 tokens

workflow-orchestrator

Manages background agent delegation, task parallelization, and result synthesis for workflow automation.

Data-Wise/claude-plugins · 21 tokens

jira-analyst

Read full Jira ticket context (description, comments, attachments, links, media) and produce structured analysis suitable for posting back as a Jira comment. Read-only via the jira-as CLI wrapper. Routed by mk:jira-analyst skill. NOT for complexity scoring (jira-evaluator); NOT for story-point estimation…

ngocsangyem/MeowKit · 72 tokens

logseq-week-info-getter

Get the logseq week days for specified or current week.

codekiln/logseq-encode-garden · 18 tokens

review-agent-workflow

Agent-workflow reviewer for AndThen review councils – catches skill-vs-agent-type confusion, gate-skipping loops, and prompt portability breaks across Claude/Codex/generic installs. Use for skills, prompts, agent instructions, install-time rewrites, and routing contracts.

IT-HUSET/andthen · 59 tokens

devops

DevOps engineer for CI/CD pipelines, Docker, GitHub Actions, infra automation, environment setup, and deployment configuration. Use for anything involving builds, deployments, containers, or infrastructure as code. Triggered by: 'CI/CD', 'GitHub Actions', 'Docker', 'deploy', 'pipeline', 'infra'.

elihuvillaraus/skills · 67 tokens