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-deliverygit 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-delivery)<a href="https://agentmods.dev/skills/thanhtung2693/specgate/specgate-work-delivery"><img src="https://agentmods.dev/badge/skills/thanhtung2693/specgate/specgate-work-delivery/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-delivery"><img src="https://agentmods.dev/badge/skills/thanhtung2693/specgate/specgate-work-delivery.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 | $0.01780 |
| Opus 5 | $0.00019 | $0.00890 |
| Sonnet 5 | $0.00008 | $0.00356 |
| Haiku 4.5 | $0.00004 | $0.00178 |
Grade A, and why
specgate-work-delivery scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 192 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Delivering Work
Apply the SpecGate operating contract. Implement the approved contract, stop at the next actor, and never approve an artifact or make a human delivery decision.
1. Load the exact contract
Use the mode from doctor. Local: read scope, criteria, verification contract,
document index, and status together:
specgate work resume "$WORK_REF" --json
Use data.status. Read pinned documents, including scope and non-goals:
specgate work context "$WORK_REF" --document "$DOCUMENT_PATH" --role "$ROLE" --json
Copy path/role from data.documents. Reuse only matching-digest content; the
index cannot replace it. Quick work uses persisted scope/criteria.
Full mode: read these; status is change status.data:
specgate work context "$WORK_REF" --json
specgate change status "$WORK_REF" --json
Stop before editing when approval is absent, the Context Pack is stale, or
criteria are missing/placeholders. Hand artifact-backed work missing its
approved version to the human or
specgate-work-preparation.
Reuse results until state changes; avoid duplicate reads. Follow status:
- If
data.next_actoris notimplementing_agent, hand off without editing. - For
review_pending, accept only onespecgatenext_commandwithout shell operators; else hand off. Run it verbatim before sections 2–5. On failure, go to section 7. On success, reread Change status; stop if stillreview_pending. - For
rework_requested, carryguidanceandmissingthrough: focused fix, affected checks, updated evidence, submit, fresh status. Never submit old evidence first. - Otherwise record
missingandguidance, then follownext_command's named section or stop if none matches.
Every change/check must map to a criterion or required repository-doc update.
Completion criterion: submitted evidence contains an observed result for every required check, or its explicit skip reason.
2. Resume safely
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.
- 2d ago Changed · -1 lines 8d1f82d9f2d9
- 8d ago First seen · 193 lines · 38 tokens per session scan A 20702c7095a9
specgate-work-delivery is a skill published in the GitHub repository thanhtung2693/specgate (5 stars, last pushed yesterday), licensed Apache-2.0. It adds 38 tokens to every session and 1,780 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
coding-agents-farm
To orchestrate parallel coding-agent farms (Claude, Codex, Copilot, Gemini, etc.) on isolated git worktrees.
delegate-coding-agent
Route coding work to another local coding-agent CLI through the llm-provider MCP server. Use when a task benefits from a powerful model for difficult reasoning or a faster, lower-cost model for bounded implementation, tests, investigation, or review.
trial
A pre-delivery evidence gate for coding agents. Hold the final response as a private draft, judge every user-visible claim against fresh covering receipts, and release only verified results. Use when unsupported completion claims, shallow green tests, stale evidence, or high-risk changes must not reach the user.
implementation-loop
Disciplined plan-build-verify loop for implementing a feature or change. Use when building something new or modifying existing behavior, especially across multiple files.
debugging-loop
Systematic reproduce-isolate-fix-verify loop for bugs, test failures, and unexpected behavior. Use when something is broken and the cause is not yet known.
trial
Trial — a pre-delivery evidence gate. Use before every final response about code or file changes: hold it as a private draft, judge every user-visible claim against fresh covering receipts, and release only verified results.