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 adam-s-k-i/super-backlog --skill task-review-gategit clone --depth 1 https://github.com/adam-s-k-i/super-backlogWrote 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/adam-s-k-i/super-backlog/task-review-gate)<a href="https://agentmods.dev/skills/adam-s-k-i/super-backlog/task-review-gate"><img src="https://agentmods.dev/badge/skills/adam-s-k-i/super-backlog/task-review-gate/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/adam-s-k-i/super-backlog/task-review-gate"><img src="https://agentmods.dev/badge/skills/adam-s-k-i/super-backlog/task-review-gate.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 | $0.00000 | $0.00544 |
| Opus 5 | $0.00000 | $0.00272 |
| Sonnet 5 | $0.00000 | $0.00109 |
| Haiku 4.5 | $0.00000 | $0.00054 |
Grade A, and why
task-review-gate 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 10d 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.
What it actually says
Task Review Gate: session entry, review gate, resume
Human checkpoint between reviewed specs and the first line of code — and the re-entry point for every session that continues a running task.
When this skill runs
- At the start of a session that works on an existing task (resume).
- Right after spec-to-backlog created tasks (review specs + acceptance criteria).
- When the user asks to implement a specific task (restate scope before starting).
Procedure
- Load the task:
backlog task view <ID> --plain. - Load the phase:
sbl phase <ID>(printsphase/spec|plan|impl|verifyornone). - Present compactly: goal, every acceptance criterion, dependencies, the recorded plan if one exists, and the current phase.
- STOP and wait for the user's explicit approval. Silence or a topic change is NOT approval.
- After approval, resume per phase:
phase/spec— review gate passed: advance withsbl phase <ID> plan, then start the plan-before-code pass.phase/plan— plan recorded: advance withsbl phase <ID> implonce the human approves the plan, then TDD.phase/impl— refresh context from the recorded plan and implementation notes, then continue TDD where it stopped.phase/verify— collect verification evidence (tests/lint/typecheck), write the final summary, thensbl phase <ID> doneat archival.none— a task without a phase label is either legacy (offersbl phase <ID> spec) or not yet started (walk the review gate first).
- Set the task In Progress via the backlog CLI when work starts.
Boundaries
- Never approve the gate yourself; vague consent is not approval.
- Phase changes only via
sbl phase— never edit labels by hand. - Trivial edits stay exempt only on explicit user instruction.
- If acceptance criteria look wrong or incomplete, send the user back to task editing instead of starting.
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.
- 10d ago Changed · +14 lines · -52 tokens per session 585cb927fd59
- 11d ago First seen · 32 lines · 52 tokens per session scan A a52587b05203
task-review-gate is a skill published in the GitHub repository adam-s-k-i/super-backlog (2 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 544 tokens. 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 2026-08-30.
Other skills, from other repositories
link-ticket-to-session
Link the current Claude Code session to a ticket (Linear, Jira, GitHub Issues, or GitHub Pull Requests) and cache its title/status in karma. Use when the user explicitly asks to link, attach, associate, or connect this session to a ticket, issue, or PR — e.g. "/link-ticket-to-session ABC-123", "link this session to…
delegate
Let Claude Architect route a versioned implementation spec through the trusted MCP runtime, independently review the Candidate Artifact, record a decision, and integrate only accepted bytes. Use for implementation delegation, Producer selection, or commitment-boundary review.
plan-review-triggers
Required reading whenever drafting any multi-step proposal, implementation plan, refactor approach, or response containing file paths and decision points. Use BEFORE writing any response that lists 3+ file paths, 2+ option-tradeoffs, a "Plan:"/"Steps:"/"Concrete changes:" section, or ends with "OK to proceed?" /…
fleet-doctrine
Fleet Deck doctrine — how to behave as one Claude Code session among many on this machine. Use when you see [FLEETDECK] messages in context, when told to coordinate with / message / answer another session, when asked about the fleet board, or when deputized to decide something for another session.
subagent-driven-delegation
Execute an implementation plan with the Superpowers subagent-driven-development loop while every task is implemented by a verified Claude Architect Producer — isolated worktree, frozen Candidate Artifact, independent verification, fresh-context review, runtime-gated acceptance. Use when executing a multi-task plan…
codex
Run the Codex CLI directly in the user's checkout for code analysis, refactoring, or automated editing without Claude Architect's verified delegation lifecycle.