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 commands/yorrick/agent-skills/dev-loopgit clone --depth 1 https://github.com/yorrick/agent-skillsWhat 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 | $0.00027 | $0.01001 |
| Opus 5 | $0.00014 | $0.00500 |
| Sonnet 5 | $0.00005 | $0.00200 |
| Haiku 4.5 | $0.00003 | $0.00100 |
Grade A, and why
dev-loop 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 2d 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 — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Development Loop
You are orchestrating a full feature development cycle.
The user's feature request is: $ARGUMENTS
Follow these phases exactly.
Phase 0: Check dependencies
Verify the script exists by running this using the Bash tool:
uv run "${CLAUDE_PLUGIN_ROOT}/scripts/dev-loop.py" --help
If uv or the script fails, tell the user they need uv installed (https://docs.astral.sh/uv/).
Phase 1: Brainstorm (interactive)
Invoke the superpowers:brainstorming skill and follow it exactly. Use the feature request above as the starting point. This is interactive — ask the user questions, explore approaches, and get design approval.
The brainstorming skill will transition to the writing-plans skill automatically. Follow that too — produce a complete implementation plan saved to docs/plans/.
Note the plan file path when done.
During brainstorming, make sure the spec includes a ## Validation section describing how to verify the feature works locally (e.g., start the server and hit an endpoint, run the CLI with specific args). This is used by the automated smoke test step after implementation.
Phase 1b: Create GitHub issue with the plan
After the plan is written and approved:
- Read the plan file content
- Create a GitHub issue using the gh CLI with the plan as the body:
gh issue create --title "" --body "$(cat )"
- Note the issue URL returned by gh. This will be passed to the script so all implementation steps reference the GitHub issue as the source of truth.
Phase 2: Hand off to automated loop
Once the issue is created, run the dev-loop orchestrator script.
Before running the script:
- Check if the current branch already has an open PR:
gh pr view --json url --jq .url - If a PR exists, tell the user and recommend using
--continue-prto implement on the current branch and review against the existing PR - Show them the GitHub issue URL
- Show the exact command that will be run
- Ask if they want to adjust --max-iterations (default 3), use --skip-permissions, or set --reviewers
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.
- 2d ago First seen · 85 lines · 27 tokens per session scan A d94986649844
dev-loop is a command published in the GitHub repository yorrick/agent-skills (10 stars, last pushed 1mo ago), licensed MIT. It adds 27 tokens to every session and 1,001 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 2026-08-31.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.