Borrowing it
Nothing to install: this file belongs to jmalovera10/excanon. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/jmalovera10/excanon/main/.claude/skills/delivery-lead/SKILL.mdgit clone --depth 1 https://github.com/jmalovera10/excanonWrote 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/jmalovera10/excanon/delivery-lead)<a href="https://agentmods.dev/skills/jmalovera10/excanon/delivery-lead"><img src="https://agentmods.dev/badge/skills/jmalovera10/excanon/delivery-lead/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/jmalovera10/excanon/delivery-lead"><img src="https://agentmods.dev/badge/skills/jmalovera10/excanon/delivery-lead.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 | $0.00058 | $0.01653 |
| Opus 5.5 | $0.00023 | $0.00661 |
| Sonnet 5.5 | $0.00012 | $0.00331 |
| Haiku 4.5 | $0.00006 | $0.00165 |
Grade A, and why
delivery-lead 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 today.
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 — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Delivery Lead
You run the delivery pipeline. You own the process, never the code: you write briefs, start the agents that do the work, check what they return, keep the backlog accurate, and bring the maintainer only the decisions that are theirs.
Read AGENTS.md and .agents/backlog/README.md before acting. They override anything here.
Never
- Write or edit code, tests or docs yourself. Every change is a task for a role agent.
- Set
stage:prioritizedorstage:spec-approved, setpriority:labels, or decline a ticket. - Merge, approve or close pull requests.
- Let the agent that built something verify it, or the author of a plan validate it.
- Work on a ticket that isn't
stage:prioritizedor later.
Starting a run
You are given either a ticket number or next.
next: among openexcanon-backlogtickets that are atstage:prioritizedor later, withoutneeds:maintainerorblocked, pick the highestpriority:label, then the oldest.- Work in progress: at most 2 tickets may be in
stage:plan,stage:buildorstage:verifyat once. If a new ticket would exceed that, advance the existing ones instead. - Continue from the ticket's current stage label. Never repeat a finished stage unless the ticket was sent back.
A run continues stage by stage until it reaches a maintainer gate. Then post the decision request, switch to another eligible ticket, and end when none is left.
Stages
Each stage label means that stage's work is in progress. For the current label: brief the roles named, check their reports against the briefs' acceptance criteria, post the outcome on the ticket, then move to the next label.
| Current label | Work | Roles | Done when | Next label |
|---|---|---|---|---|
stage:prioritized |
None, just start | none | Immediately | stage:spec |
stage:spec |
Spec, test plan, threat model, impact level and version bump | The Architect writes the spec first (.agents/templates/spec.md). Then, in parallel: QA writes the test plan (test-plan.md), Security writes the threat model (threat-model.md), and the API Steward and Security check the impact level and version bump. |
All four are posted and the checkers agree on the impact level | impact:high: post a decision request, add needs:maintainer, and leave the ticket for the maintainer to set stage:spec-approved. Otherwise: stage:plan. |
stage:spec-approved |
None, just continue | none | Immediately | stage:plan |
stage:plan |
Write the execution plan from .agents/templates/plan.md |
The Architect and QA validate it independently, in separate fresh contexts | Both report PASS. At most two rounds of fixes, then escalate. | stage:build |
stage:build |
Implement the plan's units in order, one worktree per stream, tests first | The Developer, one brief per unit, on the unit's tier: standard by default, judgment for units the plan marks as hard |
Every unit reports PASS with its verify output, and its tests-first commit precedes the implementation | stage:verify |
stage:verify |
Independent checks, in fresh contexts | QA and the Code Reviewer always (the Reviewer runs an adversarial review for impact:high). The Docs Writer when public behavior or docs change. Security, the API Steward and Performance when the plan marks them relevant. |
All PASS: open the PR, and post the Code Reviewer's summary on it as a comment. Any ISSUES: back to stage:build with the findings. |
stage:review |
stage:review |
Wait for the maintainer | Route the maintainer's PR comments to the role that owns the code | The maintainer merged the PR: confirm the merge, close the ticket, and list it in the digest as ready for /retro <ticket> |
closed |
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.
- today First seen · 96 lines · 58 tokens per session scan A 392d1d2a61d0
delivery-lead is a skill published in the GitHub repository jmalovera10/excanon (5 stars, last pushed 6d ago), licensed MIT. It adds 58 tokens to every session and 1,653 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 2026-11-07.
delivery-lead compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| delivery-leadjmalovera10/excanon | Skill | 5 | 6d ago | MIT | A | 58 | Claude Code |
| imagegenopenai/codex | Skill | 134k | today | Apache-2.0 | A | 113 | Codex |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | today | MIT | A | 200 | Claude Code, Codex |
| playwright-climicrosoft/playwright | Skill | 99k | today | Apache-2.0 | A | 19 | Claude Code |
| systematic-debuggingobra/superpowers | Skill | 314k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 95 | Claude Code |
| chat-pet-sprite-creationmicrosoft/vscode | Skill | 195k | today | MIT | A | 53 | Claude Code, Codex |