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 laurigates/claude-plugins --skill blueprint-autopilotgit clone --depth 1 https://github.com/laurigates/claude-pluginsWrote 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/laurigates/claude-plugins/blueprint-autopilot)<a href="https://agentmods.dev/skills/laurigates/claude-plugins/blueprint-autopilot"><img src="https://agentmods.dev/badge/skills/laurigates/claude-plugins/blueprint-autopilot/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/laurigates/claude-plugins/blueprint-autopilot"><img src="https://agentmods.dev/badge/skills/laurigates/claude-plugins/blueprint-autopilot.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.01921 |
| Opus 5.5 | $0.00015 | $0.00768 |
| Sonnet 5.5 | $0.00008 | $0.00384 |
| Haiku 4.5 | $0.00004 | $0.00192 |
Grade A, and why
blueprint-autopilot 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 6d 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 — 147 lines — stays where its author put it; the contents beside it link to each section on GitHub.
blueprint-autopilot
The level-2 executor of the ADR-0020 autonomy model: runs due
agent-judgment maintenance tasks and (optionally) drafts work-order
proposals — quietly, with a one-line receipt. The inverse of
/blueprint:execute: no menus, no questions, act then report.
When to Use This Skill
| Use this skill when... | Use /blueprint:execute instead when... |
|---|---|
The session-start drift nudge says blueprint tasks are due (→ /blueprint:autopilot) |
The user wants an interactive "what's next?" menu |
automation.autonomy_level ≥ 2 and ambient maintenance should just happen |
The user wants to drive each action themselves |
| A work-order proposal should be drafted from a ready PRP without interrupting the user | The user wants to create a real work order now (/blueprint:work-order) |
Context
- Automation config: !
bash ${CLAUDE_SKILL_DIR}/../../scripts/get-automation-config.sh - Due tasks (report mode, no writes): !
bash ${CLAUDE_SKILL_DIR}/../../scripts/blueprint-autorun.sh --report
Execution
Execute this autopilot pass:
Step 1: Gate on autonomy level
Read AUTONOMY_LEVEL from the Context automation config. If it is below
2, stop immediately with the single line
Autopilot inactive (autonomy_level N) — raise automation.autonomy_level in docs/blueprint/manifest.json to enable.
Nothing else. Also stop (silently, one line) when BLUEPRINT_AUTORUN_DISABLE=1
is set in the environment.
Step 2: Run due agent-judgment tasks (budget: 2 per pass)
From the Context --report output, collect the DUE_AGENT_TASKS list. Take
at most the first 2 (session token budget — the rest surface again next
session). For each task, dispatch ONE subagent (Task tool) with a focused
brief:
adr-validate→ run the/blueprint:adr-validate --report-onlyprocedure; return the findings summary onlyfeature-tracker-sync→ run the full-sync reconciliation in report-then-apply form, applying only unambiguous updates (drain evidence-backed completed WOs, recompute stats); ambiguous discrepancies are reported, not resolvedstory-audit→ run the audit and write only its dated artifact underdocs/blueprint/audits/- any other task → run its skill's
--report-only/dry-run form; never its mutating form
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.
- 6d ago Changed · +3 lines 30a130d99dbc
- 1mo ago First seen · 144 lines · 38 tokens per session scan A df02648b2023
blueprint-autopilot is a skill published in the GitHub repository laurigates/claude-plugins (60 stars, last pushed yesterday), licensed MIT. It adds 38 tokens to every session and 1,921 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 2026-09-03.
Other skills, from other repositories
morning-brief
Operator's read-only morning view of the current GitHub repo in one pass: open counts per queue label, merge-ready PRs, parked decisions with their RECOMMENDED lines, and loop-lane telemetry freshness; never mutates issues, PRs, labels, or comments. Use when: 'morning brief', 'ops dashboard', 'what needs attention'…
sdlc
Run an artifact-driven software lifecycle where each stage commits a markdown artifact that triggers the next - intent.md, spec.md, plan.md, verified diff, reviewed PR, monitored production. Use when the user asks to capture an idea as intent, write a spec, plan before implementing, set up a verification loop, define…
podio-ticket-updates
Sync in-progress Podio Kanban tickets with progress from their linked GitHub repos — propose description updates and progress comments, apply after approval. Use when the user asks to update Podio tickets, sync ticket progress, or post status updates to the Kanban board.
parallel-orchestrator
Manage parallel Claude Code workstreams using git worktrees. Use when: splitting large tasks across multiple workers, coordinating parallel development, monitoring worker progress, integrating completed work, analyzing work item documents (code reviews, issue lists). Triggers: parallel, orchestrator, worktrees…
tracker-writing
Write Jira issues, Jira comments, and GitHub pull request titles and bodies for the openEHR specification trackers and repos. Use whenever text is about to be posted to Jira or GitHub: create or edit an issue, comment, open a PR, write a review comment. Hard length caps, a fixed structure per text type, ask before…
kanban-dev-lane
Use when a Hermes Kanban worker wants to run an autonomous implementation delegation lane with a 3-tier fallback chain across claudy -> codex (--yolo) -> agyd (agy --dangerously-skip-permissions) -> Hermes direct on quota exhaustion.