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 agentmods add skills/parcadei/continuousclaudev4.7/autonomousnpx skills add parcadei/ContinuousClaudeV4.7 --skill autonomousgit clone --depth 1 https://github.com/parcadei/ContinuousClaudeV4.7Wrote 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/parcadei/continuousclaudev4.7/autonomous)<a href="https://agentmods.dev/skills/parcadei/continuousclaudev4.7/autonomous"><img src="https://agentmods.dev/badge/skills/parcadei/continuousclaudev4.7/autonomous.svg" alt="Measured on agentmods" 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 | $0.00013 | $0.03102 |
| Opus 5 | $0.00006 | $0.01551 |
| Sonnet 5 | $0.00003 | $0.00620 |
| Haiku 4.5 | $0.00001 | $0.00310 |
Grade A, and why
autonomous 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 5d 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 — 283 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Orchestrate. Never implement. Workers build; you plan, decompose, delegate, validate, steer. Carry: assertion status, file paths, contract state, pass/fail. No file operations yourself.
Pipeline: ASSESS → PLAN → PREPARE → EXECUTE → VALIDATE → EVOLVE All phases run. Depth scales with complexity.
ASSESS
Read task. Check project: tldr structure, CLAUDE.md, readiness.sh. Record baseline score. Classify: patch — bug fix, one worker, one assertion, no milestones feature — one milestone, multiple atomic workers, validation gate multi-feature — multiple milestones, validation gates between each greenfield — milestone 0 (type: approval) for design if aesthetic, then build milestones Flag aesthetic if: UI, visual, color, layout, typography. Encode as approval milestone in PLAN.
PLAN
Validation contract first — mission-level TDD. Done defined before code exists.
contract.json — single file, tracks full lifecycle:
{ "task": "Add user authentication", "complexity": "feature", "milestones": [ {"name": "auth", "status": "pending", "assertions": ["VAL-001", "VAL-002"]} ], "assertions": [ {"id": "VAL-001", "type": "invariant", "text": "Auth tokens never in logs", "milestone": "auth", "status": "pending", "depends": [], "worker": null, "evidence": null}, {"id": "VAL-002", "type": "behavioral", "text": "Login redirects to dashboard", "milestone": "auth", "status": "pending", "depends": ["VAL-001"], "presentation": "agent-browser", "worker": null, "evidence": null}, {"id": "VAL-003", "type": "approval", "text": "User approves design direction", "milestone": "design", "status": "pending", "depends": [], "presentation": "variants", "variants": 3, "medium": "agent-browser", "worker": null, "evidence": null} ] }
Lifecycle — who updates contract.json: PLAN creates it (all assertions pending) EXECUTE sets worker field on assertions it dispatches Workers never touch contract.json — they write reports/ only VALIDATE reads reports, flips assertion status to passed/failed, fills evidence EVOLVE reads final state, feeds corrections to bloks
Types: invariant (test), behavioral (e2e), contract (schema), property (prop test), fuzz (fuzz harness), approval (human — AskUserQuestion, agent-browser, or screenshot).
Aesthetic work: encode as milestone with type: approval assertions. Generate N variants in parallel, present via medium, user picks or gives direction. Loop until approval passes. Chosen direction becomes context for subsequent workers. Not a separate phase — just a milestone that gates on human approval before implementation proceeds.
Decomposition: one task = one assertion. If "and" needed to describe work, two tasks. Ask user which stages to approve. Encode as type: approval. Assertions with depends: [] run first. Assertions depending on others wait. Write: continuum/autonomous/{task-id}/contract.json Multi-milestone also: continuum/autonomous/{task-id}/plan.md
PREMORTEM
Run /premortem on contract.json + plan.md. Present findings to user via AskUserQuestion. BLOCK or WARN: user confirms before proceeding. PASS: continue to PREPARE.
PREPARE
Front-load. Workers never discover what you already know. Run these commands and CAPTURE their stdout into variables: BLOKS_CONTEXT=$(bloks context .) → project rules, tastes, corrections BLOKS_RECIPE=$(bloks recipe {lib} {keywords}) → task-specific API docs + user recipes BLOKS_CARDS=$(bloks card {lib} {symbol}) → symbol-level signatures + gotchas TLDR_STRUCTURE=$(tldr structure {path}) → affected module structure CLAUDE_CONVENTIONS from CLAUDE.md ouros session if research needed → returns compact card previous worker reports for sequential deps
These variables are LITERALLY pasted into each worker prompt CONTEXT section below. Do not summarize or rewrite bloks output. Paste verbatim. The cards are already compressed. If bloks returns empty (no cards for this lib), skip that variable — don't fill it manually.
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.
- 5d ago First seen · 283 lines · 13 tokens per session scan A 5a043f60c96e
autonomous is a skill published in the GitHub repository parcadei/ContinuousClaudeV4.7 (48 stars, last pushed 4mo ago), licensed MIT. It adds 13 tokens to every session and 3,102 once invoked, about $0.0001 per session on Opus 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 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…