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 commands/helgdemidov/refigure/feature-workflowgit clone --depth 1 https://github.com/HelgDemidov/refigureWrote 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/commands/helgdemidov/refigure/feature-workflow)<a href="https://agentmods.dev/commands/helgdemidov/refigure/feature-workflow"><img src="https://agentmods.dev/badge/commands/helgdemidov/refigure/feature-workflow.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 | $0.00018 | $0.01086 |
| Opus 5 | $0.00009 | $0.00543 |
| Sonnet 5 | $0.00004 | $0.00217 |
| Haiku 4.5 | $0.00002 | $0.00109 |
Grade A, and why
feature-workflow 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.
How it starts
The opening of the file, as written. The whole thing — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pick up a spec (ТЗ) drafted by /tech-spec that hasn't been implemented yet, and implement it end-to-end on its feature branch.
Step 1 — Find the target spec
If invoked with an argument (slug, branch name, or path), use that spec directly.
Otherwise scan docs/**/*.md (recursive — specs live under a docs/<category>/<slug>/ subdirectory, not flat, since 2026-08-05) for specs that are BOTH: missing a ## Статус выполнения section, AND whose declared branch (the **Ветка:** line) still exists — local or origin/ — and isn't merged into main. The **Ветка:** line is what distinguishes an actual spec from a general design doc (like docs/project-meta/v1-scope-and-api-design/v1-scope-and-api-design-2026-08-04.md), which has none. This two-signal check avoids false positives on old completed specs whose status section was never backfilled.
This is a plain filesystem scan (unaffected by docs/ being gitignored — untracked files are still on disk), but it only ever finds specs written by /tech-spec in THIS local checkout: since 2026-08-19 the spec file itself is never committed or pushed, so a spec drafted on a different machine, or before a fresh clone, simply isn't there to find.
- Zero matches → report nothing pending, stop.
- One match → proceed with it.
- Multiple matches → ask the user which one (AskUserQuestion) — don't guess.
Step 2 — Enter the branch
git fetch origin, then check out the spec's branch locally. /tech-spec Step 7 creates this branch but deliberately does NOT push it (no commit exists yet to push) — expect a local-only branch here, not an origin/<branch> to track. It gets pushed for the first time as part of Step 4/5 below, once real work lands on it.
Step 3 — Build the task list
Read the spec's "План коммитов/PR" section and turn it into a task list in this conversation, one task per planned commit.
If the spec has no commit breakdown (only a final-PR description), split the work into commits yourself — one coherent, independently buildable unit of change per commit (natural seams: a single module ported, its tests, the public-API wiring, docs).
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.
- 3d ago First seen · 44 lines · 18 tokens per session scan A 72a45df38b9b
feature-workflow is a command published in the GitHub repository HelgDemidov/refigure (0 stars, last pushed 6d ago), licensed Apache-2.0. It adds 18 tokens to every session and 1,086 once invoked, about $0.0001 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 commands, from other repositories
claude-cmd-status
A Claude Code slash command. OpenCode does not read Claude commands.
queue
Manage the work queue. Say "what's on my plate", "what's moving", or "add to queue: " (or run /founder-os:queue). Shows ACTIVE items, adds to BACKLOG, moves items between states. ACTIVE is capped at 3.
speckit.memory-md.log-finding
Turn a high-signal audit finding into a tracker-ready follow-up for GitHub, GitLab, Jira, or another issue system.
whatneedsdoing
One honest "what needs doing" across the two work substrates — repo dev work (docs/backlog/ + open gripes + open GitHub PRs + Dependabot alerts) and the prod factory queue (open/doable todos) — plus a repo-hygiene scan (migration-number collisions · backlog lint · memory-index lint), a prod system-health read…
project-recap
Generate a visual project recap for context switching.
brand-generate
Generate an on-brand document from a saved Brand Profile.