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 donald-ada/workinggenius --skill triagegit clone --depth 1 https://github.com/donald-ada/workinggeniusWrote 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/donald-ada/workinggenius/triage)<a href="https://agentmods.dev/skills/donald-ada/workinggenius/triage"><img src="https://agentmods.dev/badge/skills/donald-ada/workinggenius/triage/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/donald-ada/workinggenius/triage"><img src="https://agentmods.dev/badge/skills/donald-ada/workinggenius/triage.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 | $0.00056 | $0.01224 |
| Opus 5.5 | $0.00022 | $0.00490 |
| Sonnet 5.5 | $0.00011 | $0.00245 |
| Haiku 4.5 | $0.00006 | $0.00122 |
Grade A, and why
triage 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 1mo 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 — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Triage
.genius/BACKLOG.md is read whole on every /genius, and grows by how many discoveries the flow made. The rules bounding it are written already — the seed's shape and where a line goes when it leaves (BACKLOG-FORMAT.md), why it stops at 300 characters (record-prose) — each firing in its own moment, with nothing going back to check. /reconcile reaches the file from the repo's side. What no pass asks is what the flow's own records, and the user, say about these lines.
The concept: the rules exist; this is the pass that enforces them, one line at a time, and the user rules on each.
When to run it
The user types it — after a stretch of slices that each dropped one, when /genius shows a backlog nobody can scan, or when a work's leftovers land. Not on a schedule: a pass sent looking finds something, and a seed argued away on a made-up reason is worse than a long file. With an argument, only the line whose subject it names, and only the two questions one line can answer — should it be here, has it outgrown a line; pairing needs candidates and order needs the file, so say those went unasked.
What gets asked
Four questions in this order, because each earlier one can take away what a later one would have worked on:
- Should it still be here at all? Two ways it should not. A work already answered it — asked against the flow's own state, not the code: does a work name this seed's subject in its Problem, Slices or contract, with that slice closed? Which move that earns turns on which came first — the format holds that line. A work whose snapshot says the thing is still not done is evidence to keep the seed. Or its reason stopped holding. No file holds that answer, so the seed's own stated why is quoted back, and the user says whether it stands.
- Are two of them one discovery? Propose the pair, both texts, never the merge. Mechanical signals — a shared anchor, a shared symbol, a seed calling itself its neighbour's illness — were all false positives on the one real file tried, so nothing merges on a signal; the reason must be checkable in the two texts. The format holds the veto and the mechanics.
- Has it outgrown a line? Over 300 characters, counted in characters and never in bytes, the way the format's instrument counts (
python3 ${CLAUDE_SKILL_DIR}/../genius-file/measure.py statussays how many seeds are past the bound) — a byte count reads every non-ASCII seed as longer than it is, so its error only ever runs toward manufacturing work.record-prosenames two cases; this tells them apart: a seed still holding an undecided choice is carrying detail that belongs elsewhere — route it out, to the log its link names or toBACKLOG.log.md, and trim. A seed that is one chosen route plus its steps is a plan, offered as work — one at a time, never a batch, because a work file is larger than the line it replaces and every status pass reads it. - What does the order say? Top to bottom is the file's statement of what to do next. This pass proposes the position and the sentence it was read from; the text itself is left to the user.
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.
- 1mo ago Changed · +7 lines 17fbcf5cdc85
- 1mo ago Changed 2f7f775ece6b
- 1mo ago Changed · -2 lines bd00ce44ca07
- 2mo ago First seen · 38 lines · 56 tokens per session scan A 56a2d509dabb
triage is a skill published in the GitHub repository donald-ada/workinggenius (6 stars, last pushed today), licensed MIT. It adds 56 tokens to every session and 1,224 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-09-05.
Other skills, from other repositories
daily-working
End-to-end pipeline: pull a task from the project's task tracker by ID (Redmine or GitHub Issues today, more addable via a new adapter), sanity-check and impact-assess it against the codebase before touching anything, implement it with the Claude CLI, verify the result in a real browser via the Claude Chrome extension…
tasknotes
This skill should be used when the user asks to 'create a task', 'add a TODO', 'plan a sprint', 'organize tasks', 'track time', 'setup Kanban', 'create task from code', 'sync tasks with sprint', 'pomodoro session', 'start pomodoro', 'time estimate', or mentions TaskNotes, task management, or project planning in…
work
Dispatch a tracker ticket into an isolated worktree and drive it autonomously to an open pull request, then stop for review. Pass interactive to brainstorm the ticket with a human before planning it, or a sentence of editorial direction as a rider. Use when asked to work a ticket end-to-end.
batch
Dispatch a night's batch of Todo tickets — one worktree and one pull request each — and leave a morning summary. Run explicitly; never inferred.
prep
Spec a ticket before it is dispatched. Brainstorms it, decides everything it can recommend an answer for, asks a present human about the little that survives, and leaves the spec, the decisions and anything still open as one ticket comment. No worktree, no branch, no pull request, and the ticket stays where it was.
dw-prune
Walk .ai/backlog/ and decide every entry out loud: drop what is stale or already done, do what is cheap and unblocked, bundle what ships with a cousin, leave the rest. Explicit-invoke only.