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 agentmods add skills/izkreny/agentifico/trackernpx skills add izkreny/agentifico --skill trackergit clone --depth 1 https://github.com/izkreny/agentificoWrote 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/izkreny/agentifico/tracker)<a href="https://agentmods.dev/skills/izkreny/agentifico/tracker"><img src="https://agentmods.dev/badge/skills/izkreny/agentifico/tracker.svg" alt="Measured on agentmods" 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 | $0.00216 | $0.02580 |
| Opus 5 | $0.00108 | $0.01290 |
| Sonnet 5 | $0.00043 | $0.00516 |
| Haiku 4.5 | $0.00022 | $0.00258 |
Grade B, and why
tracker scanned grade B with 1 finding 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 4d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
Optional. If `.agents/gh-solo.md` exists in the repository, read it: it records the label taxonomy and layer set, the branch format and its `{type}` vocabulary, whether the repo uses GitHub issue types, the default branc How it starts
The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tools used:
Bash(gh:*)for every tracker read and write,Bash(git:*)for branch context,Readfor the standards and the per-repo config,Writefor issue body files passed to--body-file.
The user invoked this skill with the argument: $ARGUMENTS
This is a routing skill. Read $ARGUMENTS and the conversation context, pick exactly one workflow, read that workflow file, and follow its instructions inline. Do nothing else.
All paths below are relative to this skill's own directory. Resolve them against wherever this skill is installed rather than assuming a location: it ships inside a plugin, so its root is whatever the harness installed the plugin to, and the working directory is the repository being tracked.
Issue, ticket and task all mean the same thing here: a GitHub issue. The owner uses the three interchangeably and expects you to. "Write tickets for X", "break this into issues" and "give me the next task" are requests about the same object, and none of them implies a different tool, a different tracker or a different level of granularity.
Every file here is written to the agent, so you is the agent reading it. The human is the owner, always in the third person. The exceptions are workflows/help.md, whose contents are printed to the owner, and README.md, which the owner reads rendered; both address them directly.
Who this is for
One person owns the repository, writes the issues and does the work. That assumption is load-bearing and it is why this skill is shorter than a team equivalent. What is absent is coordination between developers: no estimate to defend to peers, no ticket handed over for someone else to pick up, no sprint boundary another developer is waiting on. Where a team process adds a gate to keep two people in sync, this one records a fact and moves on. Who else may be around it turns on one test: whether they write to the repository.
The team habit it keeps is assignment, repurposed: @me means work has started on this. A GitHub issue has only open and closed, and assignment supplies the missing middle. Assign when work starts, unassign when it is set aside unfinished. It deliberately does not mean queued: the only operation that assigns is Start work, which cuts the branch in the same breath, so nothing in this skill can produce a queued issue and a wider meaning would describe a state the tooling cannot reach. Closing needs no unassign: the issue keeps @me and drops out of the list anyway, because every view that matters filters on open. The one place that bites is reopening, which brings the old assignee back with it. assignee:@me is:open is then a live answer to "what am I in the middle of", across every repository at once.
Ownership is not what it records. Every issue in the owner's own repository is implicitly theirs, so assigning on creation would say nothing and cost the signal its meaning.
Pull requests follow the same rule: whoever opens one is assigned to it. It needs saying because GitHub will not do it, recording the author but leaving the assignee empty, so it takes an explicit --assignee @me at creation. The scheme is consistent, not two rules: assigned means in flight. An issue is only in flight while the owner is working on it, so it is assigned and later cleared; a PR is in flight for its whole life, so it is assigned once at creation and needs no clearing. This skill does not open pull requests itself — pr-flow does, and it sets the assignee at creation — but the rule holds wherever one is opened.
The test is write access, not headcount: the owner owns the repository and is the only person who writes to it. Other people may be involved, and it changes nothing as long as the tracker and the code have one author.
- A client who sets scope, priorities and deadlines: fine. They decide what matters, the owner still writes every issue.
- A mentor or reviewer who reads issues, comments on them and reviews pull requests: fine. Comments are not writes, and nothing in these standards waits on their approval.
- A second committer, a QA sign-off that gates a merge, or anyone else who creates or edits issues: out of scope. Stop and raise it rather than adapting these standards to fit, because they will quietly under-serve it.
What ships with it
9 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 4d ago First seen · 91 lines · 216 tokens per session scan B fb57f2d78ddf
tracker is a skill published in the GitHub repository izkreny/agentifico (2 stars, last pushed 4d ago), licensed MIT. It adds 216 tokens to every session and 2,580 once invoked, about $0.0011 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…