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.
git 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/commands/theam/claude-dev-kit/delegate-backlog)<a href="https://agentmods.dev/commands/theam/claude-dev-kit/delegate-backlog"><img src="https://agentmods.dev/badge/commands/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/commands/theam/claude-dev-kit/delegate-backlog"><img src="https://agentmods.dev/badge/commands/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 | $0.00055 | $0.00540 |
| Opus 5.5 | $0.00022 | $0.00216 |
| Sonnet 5.5 | $0.00011 | $0.00108 |
| Haiku 4.5 | $0.00006 | $0.00054 |
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 17d 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
Delegate the approved backlog referenced in the arguments to an async coding agent: $ARGUMENTS
Follow the delegate-backlog skill. This is the async, throughput-first alternative to /work-story — it fans an already-approved backlog out to GitHub Copilot coding agent (one branch/PR per issue) and then re-applies the kit's quality via pr-review/fix-pr on the PRs that come back.
Rules:
- Approved backlog only. Delegate the issues the user approved (usually
plan-backlog's output). If none is referenced, ask — don't invent scope. - GitHub only in v1 (Copilot coding agent). If the tracker isn't GitHub, say async delegation isn't available for it yet.
- Be honest about the trade-off up front: the kit's in-session gates (plan approval, coverage/security/e2e) don't run on the delegated path — they're re-applied on review. Never present a delegated PR as if the kit's gates produced it.
Run it
- Approval gate — show which issues go to which agent and the trade-off; delegate nothing until the user says yes.
- Enrich each issue with a self-contained brief (AC, repo conventions, test commands, branch naming, provenance). You may delegate the discovery/enrichment groundwork to the
backlog-delegatorsubagent. - Verify Copilot coding agent is enabled (query the repo's assignable actors) — never fabricate a trigger; if it isn't enabled, stop and say how to enable it.
- Delegate each issue by assigning it to Copilot (verify the current mechanism), confirm by read-back, and capture the resulting draft PR from the issue's linked PRs — never invent a PR number.
- Track issue → branch → PR in a run manifest (a comment on the parent/epic issue).
- Review loop — as draft PRs appear, run
pr-reviewthenfix-pron each (respecting fix-pr's open-PR guard); record verdicts in the manifest. - Report from the manifest with provenance, and never auto-merge — merging is the user's explicit per-PR call.
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.
- 17d ago First seen · 24 lines · 55 tokens per session scan A 039623e8bcd3
delegate-backlog is a command published in the GitHub repository theam/claude-dev-kit (15 stars, last pushed 18d ago), licensed Apache-2.0. It adds 55 tokens to every session and 540 once invoked, about $0.0002 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 commands, from other repositories
octo-schedule
Manage scheduled workflow jobs for the Claude Octopus scheduler.
octo-scheduler
Manage the Claude Octopus scheduled workflow runner daemon.
next
Read the active sprint’s current status. Refresh stale compilation using grainulator compile --dir or grainulator.compile. Use the resulting nextactions with the user’s current scope and authorization. Output only the two Auto and Manual lists below, with no recap, status counters or offer to continue. Continue useful…
specmanager-build
Build one phase of a SpecManager feature's plan via the builder subagent. Stops at the phase boundary; never advances.
standup
Show a daily standup summary with completed, in-progress, and blocked tasks across all active epics.
prd-score
Score an existing PRD against the 100-point AI-optimization framework.