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 thanhtung2693/specgate --skill specgate-work-preparationgit clone --depth 1 https://github.com/thanhtung2693/specgateWrote 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/thanhtung2693/specgate/specgate-work-preparation)<a href="https://agentmods.dev/skills/thanhtung2693/specgate/specgate-work-preparation"><img src="https://agentmods.dev/badge/skills/thanhtung2693/specgate/specgate-work-preparation/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/thanhtung2693/specgate/specgate-work-preparation"><img src="https://agentmods.dev/badge/skills/thanhtung2693/specgate/specgate-work-preparation.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 | $0.00038 | $0.01703 |
| Opus 5 | $0.00019 | $0.00851 |
| Sonnet 5 | $0.00008 | $0.00341 |
| Haiku 4.5 | $0.00004 | $0.00170 |
Grade A, and why
specgate-work-preparation 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 9d 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 — 205 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Preparing Work
Apply the SpecGate operating contract. This phase produces an approved implementation handoff; it never implements product code.
1. Define the contract and route
Read the request, governing repository instructions, and author-selected source
documents. Show the human the exact title, description, observable acceptance
criteria, and non-goals. Split work that can be accepted independently. Use an
@check:<name> binding only when the human confirms that exact check, and name
in the preview which criteria a check enforces and which rest on your claim.
Choose one route with the human:
- Quick work for a small change that does not need a governed source snapshot.
- Artifact-backed work when an existing spec, design, plan, verification document, or other source must be versioned and approved.
Do not create a durable record until the human approves the displayed contract.
Completion criterion: the human-approved preview names every slice's contract and route.
2A. Create quick work
Quick work is available in Local and Full mode. Persist the approved contract with explicit criteria, then read it back:
specgate work create-quick "$TITLE" --description "$DESCRIPTION" \
--ac "$CONFIRMED_CRITERION_1" --json
specgate work show "$WORK_REF" --json
specgate work context "$WORK_REF" --json
Never derive criteria from filenames, headings, numbering, or keywords. If the persisted title, description, or ordered criteria differ from the preview, stop instead of implementing.
Completion criterion: the returned Context Pack reproduces the approved quick
contract exactly. Quick work ends here; switch to specgate-work-delivery.
2B. Preview and publish an artifact
The originating framework owns source paths, names, lifecycle, and Git policy. Never relocate, copy, rename, delete, commit, or change their ignore rules. Edit source content only when the request authorizes it.
Roles are routing labels, not one-file-per-concern requirements. Map each
human-selected source explicitly; never detect a framework or manufacture
documents. When policy requires a role one source already covers,
map that same path again under the second role — one spec is often both spec
and plan.
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.
- 9d ago First seen · 205 lines · 38 tokens per session scan A e4001eafc3c9
specgate-work-preparation is a skill published in the GitHub repository thanhtung2693/specgate (5 stars, last pushed 2d ago), licensed Apache-2.0. It adds 38 tokens to every session and 1,703 once invoked, about $0.0002 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
issue-creation
Trigger: issue creation, bug reports, feature requests, or issue approval. Create and triage GitHub issues from repository evidence.
sdd-tasks
Break an SDD change into implementation tasks. Trigger: orchestrator launches task planning for a change.
systemic-issue-triage
Trigger: new issue, bug report, triage, backlog, issue flood, community report, root cause, dead-end, blocked user. Attack issues by root class, never one-by-one; fixes must shrink the system, not grow it.
init-workspace-flow
Workflow for initializing or upgrading a workspace: context, discovery, documentation, etc.
arrange-workspace-flow
Workflow for arranging a workspace: layout, reference source code, business/technical context, ecosystem setup.
ijfw-workflow
Use when the user says: 'build', 'create', 'plan', 'new project', 'brainstorm', 'design', 'UI', 'website', 'dashboard', 'app', 'help me build', 'launch', 'book', 'campaign', or anything project-level. Skill body decides Quick vs Deep path.