pyrex41/pi-extensions/shen-backpressure

Claude Code and agent skill by pyrex41/pi-extensions

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Shen specifications, guard constructors, verification gates and backpressure in explicitly configured SB projects.

Install

npx vskill@latest install pyrex41/pi-extensions/shen-backpressure

SKILL.md

Shen-Backpressure for Pi

Use sb op context to inspect the project's actual spec, output path, proof chain and gate pipeline. Do not assume specs/core.shen, fixed constructor names, or a fixed gate count. Do not scrape internal CLI logs for gate status.

SB requires a trusted project with sb.toml at cwd or an ancestor within the repository. [paths] spec and output must be explicit and the spec must exist. Nested sketches are not activated automatically. Report missing configuration; do not present convention-derived defaults as a live project.

init requires human confirmation (or the user's explicit headless --allow-sb-init flag). It skips Claude skills. Never initialize without approval.

Workflow

  1. Wrap raw inputs with generated constructors at boundaries; follow the actual proof chain.
  2. Never bypass constructors, ignore their errors, or hand-edit generated guards.
  3. Change the real spec and run gen when regeneration is needed.
  4. Run gates after changes. Fix failures before continuing or completing the task.
  5. derive checks equivalence; intentional rewrites use regen: true. audit shows the SB discharge report.

Only gates establishes passing evidence. Successful context/audit commands cannot clear a failed gate state. Pi blocks direct edit/write of the configured generated output and requires fresh gate evidence before SCUD done/commit. File changes, generation, shell calls and reload invalidate evidence. These checks do not sandbox arbitrary shell commands or external writers, and do not prove facts beyond the configured spec/TCB/gates.

Commands

  • /sb: context report, retained in the transcript/model history.
  • /sb-gates: run gates and retain the bounded report without starting a model turn.
  • /sb-fix: prompt the agent to investigate and fix backpressure (distinct from the command).
  • /sb-cancel: cancel an active slash command; Escape also cancels in the TUI.

Outputs are bounded to a 24 KiB/950-line tail. Keep credentials out of gate output. See the package README for fingerprint coverage, concurrency and execution limits.

Headless Ralph loops remain on sb loop; explicitly configure RALPH_HARNESS="pi -p" when desired. The extension does not silently change a manifest's harness or migrate another project's configuration.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
5

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/pyrex41/pi-extensions/shen-backpressure/badge)](https://verified-skill.com/skills/pyrex41/pi-extensions/shen-backpressure)

Something wrong with this skill or its page? Tell us.