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 theam/claude-dev-kit --skill delegate-backloggit clone --depth 1 https://github.com/theam/claude-dev-kitWrote 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/theam/claude-dev-kit/delegate-backlog)<a href="https://agentmods.dev/skills/theam/claude-dev-kit/delegate-backlog"><img src="https://agentmods.dev/badge/skills/theam/claude-dev-kit/delegate-backlog/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/theam/claude-dev-kit/delegate-backlog"><img src="https://agentmods.dev/badge/skills/theam/claude-dev-kit/delegate-backlog.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 | $0.00107 | $0.01846 |
| Opus 5.5 | $0.00043 | $0.00738 |
| Sonnet 5.5 | $0.00021 | $0.00369 |
| Haiku 4.5 | $0.00011 | $0.00185 |
Grade A, and why
delegate-backlog 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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Delegate Backlog
Fan an already-approved backlog out to a real async coding agent so several stories are built in parallel — one branch, one PR per issue — and then re-apply the kit's quality on the PRs that come back. This is the async, throughput-first alternative to work-story (which works one story at a time, interactively, in your session).
Experimental (v1). GitHub + Copilot coding agent only, and it leans on a young GitHub surface (§3–§4) — verify the mechanism is current and expect it to change.
Be honest about the trade-off. On the delegated path the kit's in-session gates do not run — the plan-approval gate, coverage/security/e2e, and stack conventions are your session's guarantees; an external agent runs its own loop. What the kit re-adds is at the end: pr-review / fix-pr on each resulting PR. So delegate-backlog trades the kit's gates for throughput, and buys the quality back on review. Say this to the user; never present a delegated PR as if the kit's gates produced it.
When to use vs. work-story
work-story <KEY>— one story, interactively, with the full in-session gates. The default; highest assurance.delegate-backlog— many stories at once, async, gates re-applied on review. Use it for parallel throughput when the user explicitly wants to fan out a backlog and accepts the trade-off.
0. Preconditions (verify, don't assume)
- An approved backlog exists. Delegate only stories the user has approved (typically the output of
plan-backlog). If there's no approved backlog, runplan-backlogfirst — don't invent scope. - The tracker is GitHub. v1 targets GitHub Copilot coding agent only. If
tracker.typeisn'tgithub, stop and say async delegation isn't available for this tracker yet (other targets are a later step). - The async target is real and enabled — see §3. Never fabricate a trigger (same doctrine as the "no fabricated automation" guard: the kit has no invented
/builder-style mechanism).
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 First seen · 83 lines · 107 tokens per session scan A 83a5df06928e
delegate-backlog is a skill published in the GitHub repository theam/claude-dev-kit (16 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 107 tokens to every session and 1,846 once invoked, about $0.0004 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 2026-09-29.
Other skills, from other repositories
configure
Explain the unfurl-context plugin to the user and show them how it's currently configured + what they can change. Use ONLY when the user explicitly invokes /unfurl-context:configure or asks to set up / configure / inspect / troubleshoot unfurl-context. Do NOT auto-trigger from generic mentions of GitHub, GitLab, Jira…
triage-task
Reads a ticket from Jira or Azure DevOps and writes .workflow/ /triage.json. Use when asked what to work on, to pick up a ticket by key, or to understand its description, comments and blocking links.
jira
Access Jira to search, summarize, and analyze tickets. Use when the user mentions a Jira ticket key (e.g. PROJ-1234), asks for a team digest, or wants to search/query Jira. Also use when the user wants to create issues, assign tickets, change ticket status, link issues, add comments, update fields, or do anything…
pm-sync-jira
Sync task state with Jira via MCP (optional, graceful degradation).
pm-plan
Break a feature or epic into 3-8 scoped tasks with dependency mapping.
pm-sync-wrike
Sync task state with Wrike via MCP (optional, graceful degradation).