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 HiAgencia/hi-claude --skill roadmapgit clone --depth 1 https://github.com/HiAgencia/hi-claudeWrote 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/hiagencia/hi-claude/roadmap)<a href="https://agentmods.dev/skills/hiagencia/hi-claude/roadmap"><img src="https://agentmods.dev/badge/skills/hiagencia/hi-claude/roadmap/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/hiagencia/hi-claude/roadmap"><img src="https://agentmods.dev/badge/skills/hiagencia/hi-claude/roadmap.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 | $0.00174 | $0.01256 |
| Opus 5 | $0.00087 | $0.00628 |
| Sonnet 5 | $0.00035 | $0.00251 |
| Haiku 4.5 | $0.00017 | $0.00126 |
Grade A, and why
roadmap 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 11d 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 — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ROADMAP protocol (hi-claude)
docs/ROADMAP.md is the single register of what is MISSING. docs/ESTADO.md (STATE.md in English
projects) is the live picture of what EXISTS. Neither keeps a log.
The nine rules
- An item is DELETED when it reaches 100% — never marked done. If closing it changed the state of
the system, that goes in-place into
ESTADO.md. The design of what was built lives in the work itself and in the commits, not in a list of trophies. - The progress line is ONE line, rewritten, never stacked.
Ya resuelto:/Resolved:carries what is already standing, so a session that arrives cold does not redo it. Past two lines, the item was several items: split it. - The open-work block holds at most 3 items, between the
hi-claude:en-cursomarkers. That block is injected at every session start and is what answers "what were we doing". More than three and it stops being an answer. - Closing route: unmarked =
[C], closable by a work session on its own. Mark only what a session cannot close alone —[V]needs the real world ·[O]needs a window, access or coordination ·[D]needs the user's GO ·[B]needs a dedicated measuring session. Marking only what saves time is the whole point of marking. (Done: …)is runnable when the item admits one — a grep that must return 0, a test that fails before and passes after, a command that must exit 0, a URL that must respond. Turning the criterion into something runnable is the FIRST move when taking an item. If it cannot be made runnable, that is information about the item: it is probably[V], not[C].- Timeless. No dates, no "today", no "this session".
hi-claude:work-protocolgoverns how every line is written — same rules as any other document. - HUBS first, classes not bullets. An item that unblocks others goes first. Twenty defects sharing one root cause are ONE job with one rule that closes them all; taking them one by one fixes the same thing three times.
- Work that is not for now lives in the HORIZON of its thematic block, under a
Horizonte/Horizonheading, and it carries the CONDITION that brings it in — not a date, the thing that has to become true. It stays in the same file: migrating is moving a block, and a second file is a second place where something can expire unseen. The horizon does not ride into the session, so it costs nothing at startup and it is not a place to hide work that already started. - The register is written so a PIECE can check it, and what that piece cannot read is a defect of
the ITEM. Three invariants, none of them tied to a language: the runnable forms live in the piece
and never in the item · what it cannot measure comes back as a THIRD state, never as "open" ·
and the count of criteria it cannot read is a ceiling that only goes DOWN.
references/roadmap-anatomy.mdcarries the contract and what it costs.
What ships with it
1 file 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.
- 11d ago First seen · 68 lines · 174 tokens per session scan A c85bd06795ca
roadmap is a skill published in the GitHub repository HiAgencia/hi-claude (2 stars, last pushed 20d ago), licensed MIT. It adds 174 tokens to every session and 1,256 once invoked, about $0.0009 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.
Other skills, from other repositories
recipe-create-meet-space
Create a Google Meet meeting space and share the join link.
workthreads
SpecStory Workthreads - a weekly work-thread rollup across a team's repos from SpecStory coding histories (any agent - Claude Code, Codex, Cursor, Gemini, and more). It groups the window's sessions into threads of work per project and labels each new / open / recently closed, so a lead sees what shipped, what is still…
atmos-config
Atmos root configuration: atmos.yaml discovery, precedence, deep merging, basepath, imports, minimal bootstrap, and routing to narrower Atmos skills.
story-readiness
Validate that a story file is implementation-ready. Checks for embedded GDD requirements, ADR references, engine notes, clear acceptance criteria, and no open design questions. Produces READY / NEEDS WORK / BLOCKED verdict with specific gaps. Use when user says 'is this story ready', 'can I start on this story', 'is…
autotask-creator
Rules for automation CRUD from the group-chat commander. The commander does not call mutation tools and does not edit cloud/autotasks files directly. It emits one or more top-level ... containers in its final text; the bus parses and applies them after the turn.
projects
List all managed projects with status, branch, open PRs, and open issue counts — portfolio-level view.